WO2021147995A1 - 智能汽车的控制方法、装置及相关设备 - Google Patents

智能汽车的控制方法、装置及相关设备 Download PDF

Info

Publication number
WO2021147995A1
WO2021147995A1 PCT/CN2021/073282 CN2021073282W WO2021147995A1 WO 2021147995 A1 WO2021147995 A1 WO 2021147995A1 CN 2021073282 W CN2021073282 W CN 2021073282W WO 2021147995 A1 WO2021147995 A1 WO 2021147995A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
acquisition
period
image acquisition
acquisition device
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2021/073282
Other languages
English (en)
French (fr)
Inventor
赖龙珍
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huawei Technologies Co Ltd
Original Assignee
Huawei Technologies Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Huawei Technologies Co Ltd filed Critical Huawei Technologies Co Ltd
Priority to EP21744507.1A priority Critical patent/EP4086874A4/en
Publication of WO2021147995A1 publication Critical patent/WO2021147995A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/09—Arrangements for giving variable traffic instructions
    • G08G1/095—Traffic lights
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/70—Circuitry for compensating brightness variation in the scene
    • H04N23/73—Circuitry for compensating brightness variation in the scene by influencing the exposure time
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/10—Image acquisition
    • G06V10/12—Details of acquisition arrangements; Constructional details thereof
    • G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
    • G06V10/147—Details of sensors, e.g. sensor lenses
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
    • G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
    • G06V20/582—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of traffic signs
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/09—Arrangements for giving variable traffic instructions
    • G08G1/0962—Arrangements for giving variable traffic instructions having an indicator mounted inside the vehicle, e.g. giving voice messages
    • G08G1/09623—Systems involving the acquisition of information from passive traffic signs by means mounted on the vehicle

Definitions

  • This application relates to the computer field, and in particular to the control method, device and related equipment of a smart car.
  • the electronic display device is periodically turned on and off during the working process.
  • the electronic display device may be in the extinguished state in the collected image, causing the smart car to perform signal identification based on the collected image.
  • the recognition result of the recognition is not consistent with the result actually displayed by the electronic display device (the result of human eye observation). Therefore, how to provide a more accurate signal identification recognition method, and then control the smart car according to the recognition result has become a technical problem to be solved urgently.
  • the present application discloses a control method, device and related equipment of a smart car.
  • the method can increase the probability that the result of the recognition according to the image collected by the image acquisition device is consistent with the result actually indicated by the signal identifier, thereby improving the result according to the recognition The accuracy and safety of controlling smart cars.
  • the present application provides a method for controlling a smart car.
  • the method includes: when the controller recognizes the display device identified by the signal, the controller calculates the target value of the collection parameter of the image collection device in the smart car;
  • the acquisition parameters are used to indicate the acquisition period, number of exposures, exposure duration, and exposure interval duration of the image acquisition device acquired by the image acquisition device; the controller sends an adjustment instruction to the image acquisition device, and the adjustment instruction instructs the image acquisition device according to the acquisition parameters in the adjustment instruction.
  • the target value adjusts the acquisition parameters; the controller adjusts the acquisition parameters according to the image acquisition device to obtain the image identification signal identification, and controls the smart car to drive according to the identified signal identification.
  • the controller calculates the target value of the acquisition parameter of the image acquisition device, adjusts the acquisition parameter of the image acquired by the image acquisition device, and increases the image acquisition device to collect one
  • the number of exposures in the image increases the probability of overlapping between the exposure time of the image acquisition device and the time the traffic light is on, thereby increasing the probability that the traffic light in the image collected by the image acquisition device is in the on state, and improving the recognition result and the result of the controller according to the image.
  • the probability that the result actually indicated by the signal identifier is consistent, thereby improving the accuracy and safety of controlling the smart car based on the result of the recognition.
  • the aforementioned controller may be a controller in a mobile data center (mobile data center, MDC) in a vehicle.
  • MDC mobile data center
  • the controller when the controller recognizes the display device identified by the signal, calculates the target value of the acquisition parameter of the image acquisition device in the smart car, including: the controller acquires the image acquisition device before adjusting the acquisition parameter For the acquired image, perform signal identification detection on the image obtained before the image acquisition device adjusts the acquisition parameters.
  • the controller calculates the target value of the acquisition parameter of the image acquisition device in the smart car; or, the controller acquires the smart car.
  • the controller calculates the target value of the acquisition parameter of the image acquisition device in the smart car, where the display device is located in the forward direction of the smart car.
  • the controller determines whether the image capture device can capture the image including the display device through the image captured by the image capture device or according to the distance between the display device and the smart car.
  • the controller starts when it is determined that the image including the display device can be captured Calculate the target value of the acquisition parameters and generate adjustment instructions to adjust the acquisition parameters of the image acquisition device. Since the adjustment instructions are used to increase the number of exposures or the frequency of image acquisition for the image acquisition device to acquire an image, the image acquisition device can acquire Adjusting only when the image is reached can reduce the workload of the image capture device.
  • the controller calculates the target value of the acquisition parameter of the image acquisition device in the smart car, including: the controller obtains the on-off period of the display device and the length of time the signal identifier lights up in one on-off period ,
  • the on-off period includes the length of time the signal mark is on and the time that the signal mark is off;
  • the controller determines the on-off period as the acquisition period after the image acquisition device adjusts the acquisition parameters, and the acquisition period is the time it takes for the image acquisition device to acquire an image ;
  • the controller adjusts the acquisition period of the acquisition parameters according to the image acquisition device, the total exposure time of the image acquisition device in one acquisition period, and the length of time the signal identifier lights up in a period of on and off.
  • the controller After determining that the image acquisition device adjusts the acquisition parameters, The number of exposures in one acquisition period; the controller determines the total exposure time in one acquisition period and the number of exposures in one acquisition period after the image acquisition device adjusts the acquisition parameters, and determines the exposure time each time the image acquisition device adjusts the acquisition parameters; And according to the acquisition period, the total exposure time and the number of exposures after the acquisition parameters are adjusted by the image acquisition device, determine the duration of each exposure interval after the image acquisition device adjusts the acquisition parameters; the controller adjusts the acquisition period after the acquisition parameters are adjusted according to the image acquisition device, The length of each exposure, the number of exposures in one acquisition period, and the length of each exposure interval generate adjustment instructions.
  • the acquisition period of the image acquisition device is adjusted to be equal to the on-off period of the display device, and the exposure interval is made shorter than the on-off period of the display device in one on-off period. It can be ensured that the exposure time of at least one of the multiple exposures of the image acquisition device in one acquisition cycle overlaps with the time when the traffic light is on, and it can be ensured that the traffic light in each image collected by the image acquisition device is in the on state.
  • the controller adjusts the acquisition period of the acquisition parameters according to the image acquisition device, the total exposure time of the image acquisition device in one acquisition period, and the length of time the signal identifier is lit in a light-off period, Determine the number of exposures in one acquisition cycle after the image acquisition device adjusts the acquisition parameters. Specifically, determine the number of exposures in one acquisition cycle after the image acquisition device adjusts the acquisition parameters according to the following formula: (T2-T 2b )/n ⁇ T Bright and n ⁇ T2/T bright .
  • n is the number of exposures in one acquisition period after the image acquisition device adjusts the acquisition parameters
  • T2 is the acquisition period after the image acquisition device adjusts the acquisition parameters
  • T 2b is the total acquisition period of the image acquisition device after adjusting the acquisition parameters.
  • the exposure time of T, and T light means the time that the signal indicator lights up in a light-off cycle.
  • the signal identifier is a traffic light
  • the controller adjusts the acquisition parameters according to the image acquisition device to obtain the image recognition signal identifier, and according to the recognized signal identifier to control the smart car to drive, including: the controller according to the image
  • the image obtained after the acquisition device adjusts the acquisition parameters recognizes the traffic light, and the recognition result is obtained.
  • the recognition result includes red light, green light, yellow light and unknown status.
  • the corresponding recognition result of each image is red light, green light, yellow light or unknown. Any one of the status; the controller determines the status of the traffic light in the image obtained after the image acquisition device adjusts the acquisition parameters according to the recognition result and the status of the traffic light determined last time.
  • the status of the traffic light includes red light, green light, and yellow light. Light, the state of the traffic light corresponding to each image is any one of red light, green light or yellow light.
  • the present application provides a control device for a smart car.
  • the device includes various modules for executing the control method of the smart car in the first aspect or any one of the possible implementations of the first aspect.
  • the present application provides a computing device including a processor and a memory; the memory is used to store instructions, the processor is used to execute the instructions, and when the processor executes the instructions, it executes any of the above-mentioned first aspect or any of the first aspects. The method described in the specific implementation.
  • the present application provides a non-transitory computer-readable storage medium.
  • the computer-readable storage medium stores a computer program.
  • the computer program When the computer program is executed by a computing device, it executes the first aspect or any specific implementation of the first aspect. The method described in the method.
  • this application provides a computer program product containing instructions, which when run on a computer, causes the computer to execute the methods described in the above aspects.
  • Figure 1 is a system architecture diagram of a smart car provided by the present application.
  • FIG. 2 is a schematic flowchart of a method for controlling a smart car provided by the present application
  • 3 is a schematic diagram of the relationship between the on-off period of the display device and the acquisition period of the image acquisition device provided by the present application;
  • Fig. 4 is a schematic diagram of a vehicle coordinate system provided by the present application.
  • FIG. 5 is a schematic diagram of the positional relationship between a vehicle coordinate system and a traffic light provided by the present application
  • Fig. 6 is a schematic structural diagram of a control device provided by the present application.
  • Fig. 7 is a schematic structural diagram of a computing device provided by the present application.
  • the working principle of the display device is first introduced.
  • the display device used to display the signal identification for example, a light emitting diode (LED) display
  • the display device will periodically turn on or off, that is, the display device will light up for a period of time, and then turn off for a period of time , And then turn on for a while and then turn off for a while.
  • the display device is lit for the same length of time each time, and the length of time that the display device is turned off is the same each time.
  • the sum of the duration of the display device being turned on once and the duration of being turned off once is called the on-off period of the display device.
  • the on-off period of the display device is 1/60 second to 20 milliseconds (millisecond, ms), that is, the on-off frequency of the display device is 50 Hz (Hertz, Hz) to 60 Hz.
  • the lighting time of the display device in a light-off period generally accounts for 10% to 20% of a light-off period. For example, if the on-off period of the display device is 20ms, the duration for the display device to turn on in one on-off period is 2ms to 4ms, and the off duration is 16ms to 18ms.
  • the signal identifier includes the prompt information displayed by the display device that is periodically turned on or off, including but not limited to traffic lights, and traffic prompts with text or symbols.
  • the image of the display device is collected by the on-board image acquisition device, and then the identification displayed by the display device is determined.
  • the image acquisition device needs to obtain an image of the display device through exposure.
  • Exposure refers to the process in which the light emitted or reflected by the object being photographed is projected onto the film or photosensitive element through the lens of the image capture device, and forms an image on the film or photosensitive element.
  • the time taken by the image acquisition device to acquire an image is an acquisition cycle.
  • An acquisition cycle includes the exposure time and the exposure interval.
  • the exposure time refers to the time from the shutter opening to the closing of the image acquisition equipment.
  • the image acquisition equipment completes one time within the exposure time. Exposure process.
  • the duration of the exposure interval refers to the length of time that elapses after the shutter of the image capture device is closed once before it is opened next.
  • the image acquisition frequency of the image acquisition device is 30, it means that the image acquisition device acquires 30 images per second, and the corresponding acquisition period for each image is 33.33ms, where the exposure time is 10ms, and the exposure interval is 23.33ms.
  • the acquisition frequency indicates the number of images collected by the image acquisition device per second.
  • machine vision-based detection and recognition systems are an indispensable part of smart cars.
  • Smart cars use on-board image acquisition equipment to collect images around the vehicle during driving and analyze the images to realize the driving process of the vehicle.
  • the perception of the surrounding environment provides reliable data for smart cars.
  • the controller of the mobile data center detects and recognizes the signs displayed on the display devices such as traffic lights, traffic signs (turn left and right), speed limit signs, etc. through the images collected by the image acquisition equipment, and then Control the driving of the smart car according to the recognition result.
  • the display device and the image acquisition device are two different devices, and the trigger time point when the display device is lit and the trigger time point of the image acquisition device exposure cannot be controlled synchronously. Therefore, it may happen that the exposure time when the image acquisition device collects the image does not overlap with the light-up time of the display device, which results in the display device in the off state in the image collected by the image acquisition device.
  • the current design of the display device is recognized by the human eye, taking into account the visual persistence effect of the human eye, the actual result of the display device is consistent with the result of the human eye recognition, although the display device is periodically lit or extinguished, However, due to the short on-off period, the result of human eye recognition is that the display device is always on.
  • the recognition result of the controller according to the image collected by the image acquisition device will be inconsistent with the actual result displayed by the display device. If the MDC controller controls the driving of the smart car according to the recognition result of the collected image, it may cause the smart car to violate Safety accidents even occurred in traffic rules.
  • the present application provides a control method of a smart car.
  • the signal identification displayed on the display device is effectively recognized, wherein the collection parameters include The collection period, number of exposures, exposure duration, and exposure interval of an image collected by the image collection device.
  • the image acquisition device on the smart car is able to acquire the signal identification image on the display device, it is necessary to increase the number of exposures for the image acquisition device to acquire an image, or to increase the acquisition frequency of the image acquisition device and the number of exposures for the acquisition of an image.
  • FIG. 1 is a system architecture diagram of a smart car provided by the present application.
  • the smart car 10 includes a controller 110, an image acquisition device 120, a communication box 130, and a positioning module 140.
  • the image acquisition device 120, The communication box 130 and the positioning module 140 are both communicatively connected to the controller 110, and the controller communicates with the cloud service data center or other devices through the communication box 130 and the network.
  • FIG. 2 is a schematic flowchart of a method for controlling a smart car provided by the present application, and the method is as follows S201 to S205.
  • the controller 110 recognizes the traffic light.
  • the controller 110 recognizing the traffic light means that the controller recognizes whether there is a traffic light in the image captured by the image capturing device 120.
  • the image capture device captures the image at the first capture frequency and sends the captured image to the controller 110 to control
  • the device 110 will perform traffic light detection on the received image to determine whether there is a traffic light in the image. If there is a traffic light, the controller 110 executes the following S202; if it does not exist, the controller 110 continues to receive the image sent by the image acquisition device 120 And carry out traffic light detection.
  • the controller 110 recognizing the traffic light refers to recognizing the distance between the traffic light and the smart car 10, and then determines whether there is a traffic light in the image collected by the image collection device 120 according to the distance.
  • the controller 110 may determine whether a traffic light can be detected in the image collected by the image collecting device 120 according to the location information of the smart car provided by the positioning module 140 and the map data.
  • the above-mentioned map data includes the location information of the traffic lights
  • the controller 110 obtains the current location information of the smart car provided by the positioning module 140 and the location information of the traffic lights on the driving path in the map data, and determines the smart car and the driving path. The distance between the nearest traffic lights.
  • the controller 110 determines that the traffic light can be detected in the image captured by the image capture device 120.
  • the foregoing preset distance may be the distance between the image capture device 120 and the target object when the target object is clearly recognized based on the image captured by the image capture device 120, such as 50 meters, 100 meters, or 200 meters, which is not specifically limited in the embodiment of the present application. .
  • the acquisition period for the image acquisition device 120 to acquire an image is the first acquisition period T1
  • the image acquisition device 120 An image is acquired through one exposure, that is, the first exposure number of the image acquisition device 120 is 1, the exposure duration of the image acquisition device 120 to acquire an image is the first exposure duration t 1b , and the exposure interval duration is the first exposure interval duration t 1g .
  • the first collection frequency of images collected by the image collection device 120 is 30, the first collection period is 33.33ms, where the first exposure duration may be 10ms, and the first exposure interval duration is 23.33ms.
  • S202 The controller 110 generates an adjustment instruction and sends the adjustment instruction to the image acquisition device 120.
  • the controller 110 acquires the display parameters of the display device displaying the traffic light after the image acquisition device 120 can acquire the image of the traffic light.
  • the display parameters include the on-off period of the display device and the number of traffic lights on during each on-off period. Lighting time.
  • the controller 110 calculates the acquisition parameters according to the display parameters of the display device, generates an adjustment instruction according to the calculated acquisition parameters, and sends the adjustment instruction to the image acquisition device 120.
  • the adjustment instruction instructs the image acquisition device 120 to adjust the current value of each acquisition parameter to the value corresponding to each acquisition parameter in the adjustment instruction, where the acquisition parameter includes the acquisition period, the number of exposures, the exposure duration, and the exposure interval duration.
  • the aforementioned adjustment instruction includes the image The second collection period T2 of the collection device 120, the second number of exposures n, the second exposure duration t 2b, and the second exposure interval duration t 2g .
  • the number of exposures indicates the number of exposures taken by the image capture device 120 to capture an image.
  • the second number of exposures n is a positive integer greater than 1, and the second number of exposures n indicates that the image capture device 120 captures an image after adjusting the capture parameters according to the adjustment instruction.
  • the second exposure time length t 2b represents the exposure time of each exposure when the image acquisition device 120 adjusts the acquisition parameters according to the adjustment instruction and requires multiple exposures to acquire an image.
  • the adjustment instruction instructs the image acquisition device 120 to adjust a single exposure in an acquisition period to n exposures according to the second number of exposures n, and adjust the duration of each exposure from the first exposure duration to the second exposure duration according to the second exposure duration. , Adjusting the duration of the exposure interval from the duration of the first exposure interval to the duration of the second exposure interval according to the duration of the second exposure interval. It should be noted that when the image capture device 120 receives the adjustment instruction, the image capture device 120 obtains an image through a single exposure, and the duration of the single exposure is the first exposure duration.
  • the second exposure duration is used as the exposure duration of the image capture device 120, and each exposure obtains a set of exposures.
  • the data is merged with the data obtained when the second exposure duration is used as the exposure duration n times to obtain an image. That is, after the image acquisition device 120 adjusts the exposure duration and the exposure interval duration according to the adjustment instruction, the image acquisition device 120 can obtain an image after fusing the data obtained by consecutive multiple exposures. Among them, data fusion is to merge the data obtained from multiple exposures, and finally obtain an image.
  • the method of data fusion in the traditional technology can be used, which is not limited in this application.
  • the controller 110 calculates the acquisition parameters according to the display parameters of the display device, it first determines to adjust the second acquisition period T2 of the image acquisition device 120 to be the same as the on-off period of the display device according to the acquired on-off period of the display device. If the controller 110 sets the total exposure duration of the image capture device 120 in a second capture period to T 2b , then the total exposure interval duration in a second capture period is T2-T 2b .
  • the number of exposures n in one acquisition period satisfies (T2-T 2b )/n ⁇ Ton , that is, n>(T2- T 2b )/T is bright
  • the total exposure duration T 2b in the second acquisition period T2 may be the same as the first exposure duration t 1b , or may be longer than the first exposure duration t 1b , or may be shorter than the first exposure duration t 1b .
  • the on-off period of the display device is 20 milliseconds
  • the duration of the traffic light in one on-off period is 2 ms
  • the off-time is 18 ms.
  • the first acquisition period of the image acquisition device 120 is 33.33ms
  • the first exposure duration of the image acquisition device 120 is 10ms
  • the first exposure interval duration is 23.33ms.
  • the exposure time of the image collection device 120 does not overlap with the time when the traffic light is on, resulting in the traffic light in the image collected by the image collection device 120 being off.
  • the exposure time of the image capture device 120 in the acquisition period corresponding to the N+1th image overlaps with the time when the traffic light is on, then the traffic light in the Nth image and the N+1th image To light up state.
  • the exposure time of the image acquisition device 120 does not overlap with the time when the traffic light is on, and the traffic light in the N+2th image is in the off state.
  • N is a positive integer greater than or equal to 1.
  • the controller 110 adjusts the first collection period of 33.33 ms of the image collection device 120 to a second collection period of 20 ms.
  • the controller 110 adjusts the total exposure duration of the image acquisition device 120 in one acquisition period to 12.5 ms, and the second exposure interval duration to 7.5 ms.
  • the controller 110 divides the 12.25ms exposure duration into 5 times, so each second exposure duration of the image acquisition device 120 is 2.5ms, and the exposure interval duration is 1.5ms.
  • the image capture device 120 opens the shutter 2.5ms each time to complete an exposure, and then opens the shutter again for 2.5ms at intervals of 1.5ms.
  • the image capture device 120 adjusts the first exposure duration to the second exposure duration, when the shutter is opened for the fifth time and After closing, the data obtained from the previous 5 exposures are merged to obtain an image, and the image is sent to the controller 110.
  • the image capture device 120 opens and closes the shutter for the fifth time, and then opens the shutter again every 1.5 ms. That is, after the image acquisition device 120 opens and closes the shutter at the 5mth time, it fuses the data obtained from the 5th exposure between the (5m-4)th to the 5mth time to obtain an image and sends it to the controller 110, where, m is a positive integer.
  • the second exposure interval is shorter than the time that the traffic light is on in one on-off period, even if the time when the traffic light is turned on and the time when the exposure ends are the same time, because the duration of the traffic light is longer than The exposure interval should be long.
  • the traffic light will still be on during the next exposure of the image acquisition device 120. Therefore, it can be ensured that the exposure time and the duration of the traffic light in an acquisition period overlap with each other.
  • the image acquisition period The overlap time between the exposure duration and the duration of the traffic light is at least 0.5ms and the longest is 2ms. It should be noted that the above examples are only used as examples and cannot be understood as specific limitations.
  • the second acquisition period of the image acquisition device 120 is adjusted to be equal to the on-off period of the display device, and the second exposure interval is made shorter than the on-off period of the display device. It can be ensured that the exposure time of at least one of the n exposures of the image acquisition device 120 in one acquisition period overlaps with the time when the traffic light is on, and the traffic light is in the lighted state in the image obtained by fusing the data of n exposures, that is, It can be ensured that the traffic light in each image output by the image acquisition device 120 is in the on state.
  • the above-mentioned method for the controller 110 to obtain the display parameters of the display device for displaying traffic lights includes but is not limited to the following three methods: the first method is that the display device is equipped with a sending module, and the sending module periodically turns on and off periods and Broadcasting is carried out for the duration of the light-on within a light-off period, and all smart cars approaching the display device can receive the data sent by the above-mentioned sending module, thereby obtaining the display parameters of the above-mentioned display device.
  • the second way is that after the controller 110 determines that the image of the traffic light can be collected, it sends an acquisition request to the display device through the communication box 130. The acquisition request is used to acquire the above-mentioned display parameters.
  • the display parameters are sent to the communication box 130.
  • the third way is that after the controller 110 determines that the image of the traffic light can be collected, the controller 110 sends an acquisition request to the cloud service data center through the communication box 130.
  • the acquisition request is used to acquire the above-mentioned display parameters.
  • the acquisition request includes the display The location information of the device.
  • the cloud service data center determines the corresponding relationship between the location information in the acquisition request and the location information stored in the cloud service data center and the display device according to the location information in the acquisition request.
  • the display device then obtains the display parameters of the display device corresponding to the location information and sends the display parameters to the controller 110.
  • the image acquisition device 120 receives the adjustment instruction, adjusts the acquisition parameters according to the adjustment instruction, acquires images according to the adjusted acquisition parameters, and sends the acquired images to the controller 110.
  • the image acquisition device 120 After the image acquisition device 120 receives the adjustment instruction sent by the controller 110, it adjusts the acquisition period, the number of exposures, the exposure duration, and the exposure interval duration of the image acquisition device 120 to acquire images with the adjusted acquisition parameters.
  • the image acquisition device 120 performs multiple exposures in one acquisition period, and then merges the data obtained from the multiple exposures to obtain an image acquired in the acquisition period, and sends the image to the controller 110. It can be understood that, in order to enable the controller 110 to recognize whether the traffic light is a red light, a yellow light, or a green light according to the image, the image collected by the image collection device 120 is a color image, such as an image in a red green blue (RGB) color mode. .
  • RGB red green blue
  • the controller 110 receives the image sent by the image acquisition device 120, recognizes the traffic light in the image, and determines the state of the traffic light.
  • the controller 110 After the controller 110 receives the image captured by the image capture device 120, it first determines the position information of the traffic light in the image.
  • the position information is the coordinate values of the four pixels in the image, for example, the coordinate values of the four pixels are respectively It is (u1, v1), (u1, v2), (u2, v1) and (u2, v2), and an initial rectangular area determined by these four pixels includes traffic lights.
  • the controller 110 determines the area of interest that includes the traffic light in the image according to the position information of the traffic light in the image, for example, uses the initial rectangular area as the area of interest, or determines four new pixel points according to the coordinates of the four pixel points.
  • the coordinates, the new rectangular area obtained after the initial rectangular area is expanded by a preset multiple as the area of interest, to prevent the initial rectangular area from not including complete traffic lights.
  • the four new pixel coordinates are (u1-1/2(u2-u1), v1-1/2(v2-v1)), (u1-1/2(u2-u1), v1+1 /2(v2-v1)), (u1+1/2(u2-u1), v1-1/2(v2-v1)) and (u1+1/2(u2-u1), v1+1/2 (v2-v1)), expand the initial rectangular area by 4 times as the area of interest, then intercept the above-mentioned area of interest as the target image, input the target image into the traffic light recognition network, and perform the analysis on the traffic lights in the target image. Recognize and get the recognition result.
  • the recognition results of the controller 110 according to the above target image include red light, green light, yellow light, and unknown state.
  • the unknown state indicates that there is no traffic light in the image or the traffic light in the image is off.
  • the controller 110 needs to determine the state of the traffic light, that is, whether the traffic light is a red light, a green light or a yellow light, and then determine whether the smart car needs to stop or slow down according to the state of the traffic light. Therefore, after the controller 110 obtains the recognition result of the current image according to the image recognition, it needs to determine the state of the traffic light in the image according to the state of the traffic light before the image, and determine the state of the traffic light as red light, green light or Any of the yellow lights.
  • the controller can identify traffic lights through the color information of traffic lights in the target image, and can also identify traffic lights in the target image through a neural network model trained to identify traffic lights, such as a convolutional neural network.
  • a neural network model trained to identify traffic lights, such as a convolutional neural network.
  • Models convolutional neural networks, CNN
  • recurrent neural network models recurrent neural network, RNN
  • the controller 110 regards the recognition result of the image as the state of the traffic light. If the recognition result of the current image by the controller 110 is an unknown state, the state of the traffic light determined before the image is used as the state of the corresponding traffic light of the current image, for example, the recognition result of the current image If it is an unknown state, the state of the traffic light before the image is green, and the green light is taken as the state of the corresponding traffic light of the current image.
  • the method for the controller 110 to determine the state of the traffic light in the image according to the recognition result of the current image and the state of the traffic light before the image is shown in Table 1 below, where any state in the table represents a red light or a green light. , Yellow light or unknown state.
  • the current state of the previous traffic light Current recognition result The state of the traffic light Any state Green light Green light Any state Yellow light Yellow light Any state red light red light Green light Unknown status Green light Yellow light Unknown status Yellow light red light Unknown status red light
  • the controller 110 after the controller 110 receives the image sent by the image acquisition device 120, it can determine the position information of the traffic light in the image by detecting the traffic light in the image, for example, the controller 110 110 may determine the position information of the traffic light in the image according to the color information and/or the shape information.
  • the controller 110 can also determine the position information of the traffic light relative to the smart car through the above-mentioned positioning module 140 and the image data, and determine the position information of the traffic light relative to the smart car according to the position information of the traffic light relative to the smart car. Then combine the conversion relationship between the vehicle coordinate system and the pixel coordinate system to determine the position information of the traffic light in the image.
  • the position information of the traffic light relative to the smart car includes the distance of the traffic light relative to the vehicle, the height of the traffic light, the length of the traffic light, and the distance of the vehicle from the edge of the road, where the height of the traffic light refers to the lower of the traffic light. The distance between the side edge and the road surface.
  • the length of the traffic light refers to the distance from the left side of the traffic light (relative to the vehicle) to the edge of the road.
  • FIG. 4 is a schematic view of the present application provides a vehicle coordinate system, the vehicle body in the plan view shown in FIG.
  • the position of the traffic light relative to the vehicle coordinate system can be determined.
  • FIG. 5 is a vehicle coordinate system and the position of the traffic light provided by the present application. Schematic diagram of the relationship.
  • the distance of the traffic light relative to the vehicle is the coordinate value x of the traffic light on the X w axis
  • the height of the traffic light is the coordinate value z of the traffic light on the Z w axis
  • the distance between the vehicle and the edge of the road is the right side of the traffic light (relative to vehicle) coordinate value y1 Y w axis
  • the coordinates of the four vertices of the rectangular area including the traffic light in the vehicle coordinate system are A(x, y1, z), B(x, y1, z+L), C(x, y2, z) and D(x, y2, z+L)
  • the coordinates of the above four coordinate points are the coordinates of the traffic light in the three-dimensional coordinate system, according to the difference between the vehicle coordinate system and the pixel coordinate system
  • the conversion relationship can determine the coordinates of the four coordinate points in the pixel coordinate system, that is, the position information of the traffic light in the image.
  • S205 The controller 110 controls the driving of the smart car according to the state of the traffic light.
  • the controller 110 After the controller 110 recognizes the state of the traffic light, it controls the driving of the smart car according to the state of the traffic light. Exemplarily, when the traffic light status is a green light, the smart car is controlled to keep running at the current speed; when the traffic light is a yellow light or a red light, the smart car is controlled to gradually decelerate to a stop.
  • the controller 110 recognizes the state of the traffic light according to the method in S201 to S205 above, and controls the driving of the smart car according to the state of the traffic light, the controller 110 cannot detect the display in the image collected by the image acquisition device 120.
  • a reply instruction is generated and sent to the image acquisition device 120, so that the image acquisition device adjusts the acquisition parameters to the values before the adjustment instruction is received.
  • the controller 110 after the controller 110 recognizes the traffic light by using the method in S201, the controller 110 cannot obtain the on-off period of the display device and the duration of the on-off period of the display device.
  • the controller 110 regards the on-off period normally used by the display device as the on-off period of the currently identified traffic light display device, and sets the duration of the light on during one on-off period commonly used by the display device as the current traffic light on and off. The length of time the light is on during the cycle. For example, a display device usually has a light-off cycle of 20ms, and a light-on time period of 2ms in a light-off cycle.
  • the controller 110 determines the acquisition parameters adjusted by the image acquisition device 120 according to the same method of determining the second acquisition period, the second number of exposures, the second exposure duration, and the second exposure interval duration in S202.
  • the controller 110 after the controller 110 recognizes the traffic light by using the method in S201, the controller 110 cannot obtain the on-off period of the display device and the duration of the on-off period of the display device.
  • the controller 110 acquires the current first acquisition period of the image acquisition device 120, if the first acquisition period is less than or equal to the preset acquisition period, that is, the number of images collected by the image acquisition device 120 per unit time is greater than or equal to the preset quantity.
  • the controller 110 determines that the image acquisition device 120 keeps the first acquisition period unchanged.
  • the controller 110 determines to change the collection period of the image collection device 120 from the first collection period Reduce to the preset acquisition period.
  • the preset collection period is the on-off period commonly used by the display device. For example, if the usual on-off period of the display device is 20 ms, the preset collection period is taken as 20 ms. At the same time, the controller 110 uses the on-off period of the display device as the on-off period of the current traffic light as the on-off period.
  • the controller 110 determines the acquisition period after the image acquisition device 120 adjusts the acquisition parameters and the length of time the display device is on during one on-off period according to the foregoing method, and determines the second acquisition period, the second number of exposures, and the second exposure duration according to the foregoing S202. And the second exposure interval is the same method to determine the acquisition parameters adjusted by the image acquisition device 120, where the first acquisition period is less than or equal to the first acquisition period when the preset acquisition period is, or the first acquisition period is greater than the preset acquisition
  • the second collection period in the period is the second collection period T2 in the calculation in S202.
  • the preset collection period can be calculated based on the statistical information of historical data, or it can be an empirical value.
  • the controller 110 recognizes the traffic light by using the method in S201, whether the controller 110 can obtain the on-off period of the display device and whether the display device is on during one on-off period Duration, the acquisition period, number of exposures, exposure duration, and exposure interval duration in the adjustment instructions generated by the controller 110 are all preset fixed values, for example, the acquisition period is 20ms, the number of exposures is 5, and the exposure duration is 2.5ms. The exposure interval is 1.5ms.
  • the image acquisition device 120 adjusts the acquisition parameters to the above-mentioned preset fixed values.
  • Fig. 6 is a schematic structural diagram of a control device for a smart car provided by the present application.
  • the control device is used in a smart car including an image capture device and a positioning module.
  • the image capture device is installed on the vehicle and is used to capture the driving of the vehicle.
  • the control device is connected with the image acquisition device, and is used to receive the image collected by the image acquisition device.
  • the control device 600 includes: a processing unit 610, a sending unit 620, and an acquiring unit 630, where:
  • the processing unit 610 is used to calculate the target value of the acquisition parameter of the image acquisition device in the smart car when identifying the display device identified by the signal; the acquisition parameter of the image acquisition device is used to instruct the image acquisition device to acquire the image acquisition period, number of exposures, Exposure duration and exposure interval duration.
  • the processing unit 610 can acquire the image collected by the image acquisition device through the acquisition unit 630, and identify the display device identified by the signal based on the image acquired by the image acquisition device, or identify the display device identified by the signal based on the position information provided by the positioning module and the distance of the signal identification .
  • the display device identified by the processing unit 610 to identify the signal identifier reference may be made to the related description of the controller 110 identifying the foregoing display device in S201.
  • the processing unit 610 When the processing unit 610 recognizes the display device of the signal identification, it determines that the acquisition parameters of the image acquisition device need to be adjusted.
  • the image acquisition device acquires the on-off period of the display device through the acquisition unit 630 and the signal identification displayed by the display device is on.
  • the light-on duration in the off period, the light-off period includes the duration of the signal indicator being on and the duration of the signal indicator being off.
  • the processing unit 610 can refer to the related description of the controller 110 calculating the collection parameters in S202 according to the method of calculating the collection parameters, which will not be repeated here.
  • the processing unit 610 generates an adjustment instruction according to the target value of the acquisition parameter, and the adjustment instruction instructs the image acquisition device to adjust the acquisition parameter according to the target value of the acquisition parameter in the adjustment instruction.
  • the sending unit 620 is configured to send an adjustment instruction to the image acquisition device.
  • control device 600 of the embodiment of the present application can be implemented by an application-specific integrated circuit (ASIC) or a programmable logic device (PLD).
  • ASIC application-specific integrated circuit
  • PLD programmable logic device
  • the above-mentioned PLD can be a complex program logic.
  • Device complex programmable logical device, CPLD
  • field-programmable gate array field-programmable gate array
  • FPGA field-programmable gate array
  • GAL generic array logic
  • the control device 600 and its various modules can also be software modules.
  • the processing unit 610 is further configured to obtain an image recognition signal identifier obtained after adjusting the acquisition parameters according to the image acquisition device, and control the driving of the smart car according to the recognized signal identifier.
  • the processing unit 610 can refer to the method of identifying the traffic light in S204 for the method of identifying the signal identifier according to the image collected by the image acquisition device, which will not be repeated here.
  • the control device 600 recognizes the state of the traffic light, it controls the driving of the smart car according to the state of the traffic light. Exemplarily, when the traffic light status is green, the smart car is controlled to drive at the current speed; when the traffic light is yellow or red, the smart car is controlled to gradually decelerate to a stop.
  • control device 600 can refer to related operations such as the controller 100 identifying signal identification, calculating collection parameters, and identifying traffic lights in the above method embodiment, which will not be described in detail here.
  • the computing device 700 at least includes a processor 710, a communication interface 720, and a memory 730.
  • the processor 710, the communication interface 720, and the memory 730 are connected to each other through a bus 740, where The processor 710 is configured to execute instructions stored in the memory 730.
  • the memory 730 stores program codes, and the processor 710 can call the program codes stored in the memory 720 to perform the following operations:
  • the controller When the controller recognizes the display device identified by the signal, it calculates the target value of the acquisition parameter of the image acquisition device in the smart car; the acquisition parameter of the image acquisition device is used to instruct the image acquisition device to acquire the image acquisition period and exposure Number of times, exposure duration and exposure interval duration;
  • the adjustment instruction instructing the image acquisition device to adjust the acquisition parameter according to the target value of the acquisition parameter in the adjustment instruction
  • the image identification signal identification obtained after adjusting the acquisition parameters according to the image acquisition device, and the driving of the smart car is controlled according to the identified signal identification.
  • the processor 710 may have a variety of specific implementation forms.
  • the processor 710 may be a central processing unit (CPU) or a graphics processing unit (GPU). 710 can also be a single-core processor or a multi-core processor.
  • the processor 710 may be a combination of a CPU and a hardware chip.
  • the aforementioned hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof.
  • the above-mentioned PLD may be a complex programmable logic device (CPLD), a field programmable logic gate array (FPGA), a general array logic (GAL) or any combination thereof.
  • the processor 710 may also be implemented by a logic device with built-in processing logic alone, such as an FPGA or a digital signal processor (digital signal processor, DSP).
  • the communication interface 720 can be a wired interface or a wireless interface for communicating with other modules or devices.
  • the wired interface can be an Ethernet interface, a controller area network (CAN) interface, or a local interconnect network (local interconnect network, LIN) and FlexRay interface.
  • the wireless interface can be a cellular network interface or a wireless LAN interface.
  • the communication interface 720 in the embodiment of the present application may be specifically used to receive image data sent by the image acquisition device, receive the on-off period sent by the display device, and so on.
  • the memory 730 may be a non-volatile memory, for example, read-only memory (ROM), programmable read-only memory (programmable ROM, PROM), erasable programmable read-only memory (erasable PROM, EPROM), Electrically erasable programmable read-only memory (EPROM, EEPROM) or flash memory.
  • ROM read-only memory
  • PROM programmable read-only memory
  • EPROM erasable programmable read-only memory
  • EPROM Electrically erasable programmable read-only memory
  • EEPROM Electrically erasable programmable read-only memory
  • the memory 730 may also be used to store instructions and data, so that the processor 710 can call the instructions stored in the memory 730 to implement the operations performed by the processing unit 610, such as the operations performed by the controller 110 in the foregoing method embodiment.
  • the computing device 700 may include more or fewer components than those shown in FIG. 7, or may have different component configurations.
  • the bus 740 may be a CAN bus or other internal bus that implements interconnection between various systems or devices in the vehicle.
  • the bus 740 can be divided into an address bus, a data bus, a control bus, and so on. For ease of representation, only one thick line is used in FIG. 7, but it does not mean that there is only one bus or one type of bus.
  • the computing device 700 may further include an input/output interface 750, and the input/output interface 750 is connected to an input/output device for receiving input information and outputting operation results.
  • computing device 700 may correspond to the control device 600 in the embodiment of the present application, and may correspond to performing the corresponding operations in the method shown in FIG.
  • the operations and/or functions are used to implement the corresponding procedures of the methods in FIGS. 1 to 5, and are not repeated here for brevity.
  • the embodiment of the present application also provides a non-transitory computer storage medium.
  • the computer storage medium stores instructions. When it runs on a processor, it can implement the method steps in the above method embodiments.
  • the processor of the computer storage medium is For the specific implementation of executing the steps of the foregoing method, reference may be made to the specific operations of the foregoing method embodiments, and details are not described herein again.
  • the foregoing embodiments may be implemented in whole or in part by software, hardware, firmware, or any other combination.
  • the above-mentioned embodiments may be implemented in the form of a computer program product in whole or in part.
  • the computer program product includes one or more computer instructions.
  • the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.
  • the computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium.
  • the computer instructions may be transmitted from a website, computer, server, or data center.
  • the computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that includes one or more sets of available media.
  • the usable medium may be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium, or a semiconductor medium.
  • the semiconductor medium may be a solid state drive (SSD).

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Signal Processing (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Vascular Medicine (AREA)
  • Traffic Control Systems (AREA)

Abstract

一种智能汽车（10）的控制方法、装置及相关设备，该方法包括：控制器（110）首先在识别到信号显示装置时，计算智能汽车（10）中图像采集设备（120）的采集参数的目标值；图像采集设备（120）的采集参数用于指示图像采集设备（120）采集图像的采集周期、曝光次数、曝光时长与曝光间隔时长；然后，向所图像采集设备（120）发送调整指令；再根据图像采集设备（120）调整采集参数后获得的图像识别信号标识，并根据识别的信号标识控制智能车（10）行驶。以此提高图像采集设备（120）采集的图像中信号标识为点亮状态的概率，提高控制器（110）根据图像所识别的结果与信号标识实际表示的结果相一致的概率，从而提高根据识别的结果对智能汽车（10）进行控制的准确性与安全性。

Description

智能汽车的控制方法、装置及相关设备
本申请要求于2020年01月23日提交中国专利局，申请号为202010077052.4，申请名称“智能汽车的控制方法、装置及相关设备”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及计算机领域，尤其涉及智能汽车的控制方法、装置及相关设备。
背景技术
近年来，智能汽车已成为车辆发展的新趋势，越来越多的汽车采用了辅助驾驶(driver assistance)、自动驾驶(automated driving)或智能网联驾驶(intelligent network driving)等系统，这类系统利用图像采集设备以及雷达等车载传感器在行驶过程中通过车载传感器等设备感知周围环境。其中，智能汽车对周围障碍物和基础设施的感知准确性影响智能汽车的规划和控制。而对电子显示装置上的信号标识(例如交通灯)的识别更是实现智能汽车自动驾驶的关键环节。
但是，电子显示装置在工作的过程中是周期性点亮与熄灭的，图像采集设备在采集图像时，采集的图像中电子显示装置可能处于熄灭状态，导致智能汽车根据采集的图像对信号标识进行识别的识别结果与电子显示装置实际表示的结果(人眼观察的结果)不一致。因此，如何提供一种更准确的信号标识的识别方法，进而根据识别结果对智能车进行控制成为亟待解决的技术问题。
发明内容
本申请公开了一种智能汽车的控制方法、装置及相关设备，该方法可以提高根据图像采集设备采集的图像所识别的结果与信号标识实际表示的结果相一致的概率，进而提高根据识别的结果对智能汽车进行控制的准确性与安全性。
第一方面，本申请提供一种智能汽车的控制方法，该方法包括：当控制器识别信号标识的显示装置时，控制器计算智能汽车中图像采集设备的采集参数的目标值；图像采集设备的采集参数用于指示所述图像采集设备采集图像的采集周期、曝光次数、曝光时长与曝光间隔时长；控制器向图像采集设备发送调整指令，该调整指令指示图像采集设备根据调整指令中采集参数的目标值调整采集参数；控制器根据图像采集设备调整采集参数后获得的图像识别信号标识，并根据识别的信号标识控制智能车行驶。
实施上述智能汽车的控制方法，在图像采集设备可以采集到信号标识时，控制器计算图像采集设备的采集参数的目标值，调整图像采集设备采集图像的采集参数，以增加图像采集设备采集一张图像时的曝光次数，提高图像采集设备的曝光时间与交通灯灯亮时间的重合概率，从而提高图像采集设备采集的图像中交通灯为点亮状态的概率，提升控制器根据图像所识别的结果与信号标识实际表示的结果相一致的概率，进而提高根据识别的结果对智能汽车进行控制的准确性与安全性。
在一种可能的实现方式中，上述控制器可以是车辆中移动数据中心(mobiledatacenter，MDC)中控制器。
在另一种可能的实现方式中，上述当控制器识别信号标识的显示装置时，控制器计算智能汽车中图像采集设备的采集参数的目标值，包括：控制器获取图像采集设备调整采集参数前获得的图像，对图像采集设备调整采集参数前获得的图像进行信号标识检测，当检测到信号标识时，控制器计算智能汽车中图像采集设备的采集参数的目标值；或者，控制器获取智能汽车与显示装置的距离，当距离小于或者等于预设距离时，控制器计算智能汽车中图像采集设备的采集参数的目标值，其中，显示装置位于智能汽车的前进方向上。
控制器通过图像采集设备采集的图像或者根据显示装置的与智能汽车的距离，确定图像采集设备是否可以采集到包含显示装置的图像，在确定可以采集到包含显示装置的图像时，控制器才开始计算采集参数的目标值并生成调整指令，进而调整图像采集设备的采集参数，由于该调整指令用于提高图像采集设备采集一张图像的曝光次数或采集图像的频率，在满足图像采集设备可以采集到图像时才进行调整，可以降低图像采集设备的工作量。
在另一种可能的实现方式中，控制器计算智能汽车中图像采集设备的采集参数的目标值，包括：控制器获取显示装置的亮灭周期以及信号标识在一个亮灭周期内点亮的时长，亮灭周期包括信号标识点亮的时长与信号标识熄灭的时长；控制器将亮灭周期确定为图像采集设备调整采集参数后的采集周期，采集周期为图像采集设备采集一张图像所用的时长；控制器根据图像采集设备调整采集参数后的采集周期、图像采集设备在一个采集周期内总的曝光时长以及信号标识在一个亮灭周期内点亮的时长，确定图像采集设备调整采集参数后，一个采集周期内的曝光次数；控制器根据图像采集设备调整采集参数后在一个采集周期内总的曝光时长以及一个采集周期内的曝光次数，确定图像采集设备调整采集参数后每次的曝光时长；并根据图像采集设备调整采集参数后的采集周期、总的曝光时长以及曝光次数，确定图像采集设备调整采集参数后每次的曝光间隔时长；控制器根据图像采集设备调整采集参数后的采集周期、每次的曝光时长、一个采集周期内的曝光次数以及每次的曝光间隔时长生成调整指令。
通过实施上述方法，将图像采集设备的采集周期调整为等于显示装置的亮灭周期，并使曝光间隔时长小于显示装置一个亮灭周期内点亮的时长。可以确保图像采集设备在一个采集周期内的多次曝光中至少有一次的曝光时间与交通灯点亮的时间有重叠，可以确保图像采集设备采集的每张图像中交通灯均为灯亮状态。
在另一种可能的实现方式中，控制器根据图像采集设备调整采集参数后的采集周期、图像采集设备在一个采集周期内总的曝光时长以及信号标识在一个亮灭周期内点亮的时长，确定图像采集设备调整采集参数后，一个采集周期内的曝光次数，具体为根据下述公式确定图像采集设备调整采集参数后，一个采集周期内的曝光次数：(T2-T 2b)/n<T 亮且n≤T2/T 亮。其中，n为图像采集设备调整采集参数后，一个采集周期内的曝光次数，T2为图像采集设备在调整采集参数之后的采集周期，T 2b为图像采集设备在调整采集参数之后一个采集周期内总的曝光时长，T 亮表示信号标识在一个亮灭周期内点亮的时长。
在另一种可能的实现方式中，信号标识为交通灯，控制器根据图像采集设备调整采集参数后获得的图像识别信号标识，并根据识别的信号标识控制智能车行驶，包括：控制器根据图像采集设备调整采集参数后获得的图像对交通灯进行识别，得到识别结果，识别结果包括红灯、绿灯、黄灯和未知状态，每张图像对应的识别结果为红灯、绿灯、黄灯或未知状态中的任意一种；控制器根据识别结果以及上一次确定的交通灯的状态，确定图像采集设备调整采集参数后获得的图像中交通灯的状态，交通灯的状态包括红灯、绿灯和黄灯，每张图像对应的交通灯的状态为红灯、绿灯或黄灯中的任意一种。
第二方面，本申请提供一种智能汽车的控制装置，所述装置包括用于执行第一方面或第一方面任一种可能实现方式中的智能汽车的控制方法的各个模块。
第三方面，本申请提供一种计算设备，包括处理器和存储器；存储器用于存储指令，处理器用于执行指令，当处理器执行所述指令时执行如上述第一方面或者第一方面的任意具体实现方式中所描述方法。
第四方面，本申请提供一种非瞬态的计算机可读存储介质，计算机可读存储介质存储有计算机程序，计算机程序被计算设备执行时执行如上述第一方面或者第一方面的任意具体实现方式中所描述方法。
第五方面，本申请提供了一种包含指令的计算机程序产品，当其在计算机上运行时，使得计算机执行上述各方面所述的方法。
本申请在上述各方面提供的实现方式的基础上，还可以进行进一步组合以提供更多实现方式。
附图说明
图1是本申请提供的一种智能汽车的系统架构图；
图2是本申请提供的一种智能汽车的控制方法的流程示意图；
图3是本申请提供的一种显示装置的亮灭周期与图像采集设备的采集周期之间的关系示意图；
图4是本申请提供的一种车辆坐标系的示意图；
图5是本申请提供的一种车辆坐标系与交通灯的位置关系示意图；
图6是本申请提供的一种控制装置的结构示意图；
图7是本申请提供的一种计算设备的结构示意图。
具体实施方式
下面结合附图对本申请所提供的智能汽车的控制方法进行详细的阐述。
为了便于理解本申请所提供的智能汽车的控制，首先介绍显示装置的工作原理。用于显示信号标识的显示装置(例如，发光二极管(light emitting diode，LED)显示屏)在通电之后，显示装置会周期性点亮或熄灭，即显示装置会点亮一段时间，然后熄灭一段时间，再点亮一段时间后熄灭一段时间。其中，显示装置每次点亮的时长均相同，每次熄灭的时长均相同。将显示装置点亮一次的时长与熄灭一次的时长之和称为显示装置的亮灭周期。通常的，显示装置的亮灭周期为1/60秒到20毫秒(millisecond，ms)，即显示装置的亮灭频率为50赫兹(Hertz，Hz)到60Hz。而一个亮灭周期内显示装置点亮的时长一般占一个亮灭周期的10％至20％。例如，显示装置的亮灭周期为20ms，则显示装置在一个亮灭周期内点亮的时长为2ms至4ms，熄灭的时长为16ms至18ms。其中，信号标识包括具有周期性点亮或熄灭的显示装置显示的提示信息，包括但不限于交通灯、带有文字或符号的交通提示。
在智能汽车中，通过车载图像采集设备采集显示装置的图像，进而确定显示装置所显示的标识。通常地，在采集过程中图像采集设备需要通过曝光获得显示装置的图像。曝光是指被拍摄物体发出或者反射的光线通过图像采集设备的镜头投射到底片或者感光元件上，在底片或者感光元件上成像的过程。图像采集设备采集一张图像所用的时长为一个采集周期，一个采集周期包括曝光时长和曝光间隔时长，曝光时长是指图像采集设备从快门打开到关闭的 时长，图像采集设备在曝光时长内完成一次曝光过程。曝光间隔时长是指图像采集设备的快门在一次关闭之后，下一次打开之前所经过的时长。示例性的，若图像采集设备采集图像的采集频率为30，表示图像采集设备每秒采集30张图像，每张图像对应的采集周期为33.33ms，其中曝光时长为10ms，曝光间隔时长为23.33ms，其中，采集频率指示图像采集设备每秒钟采集的图像的数量。
在智能汽车中，基于机器视觉的检测与识别系统是智能汽车中不可或缺的部分，智能汽车通过车载图像采集设备采集车辆行驶过程中周围的图像并对图像进行分析，以实现对车辆行驶过程中周围环境的感知，为智能汽车提供可靠数据。例如车载移动数据中心(mobile data center，MDC)的控制器通过图像采集设备采集的图像检测并识别交通灯、交通标志(左转与右转)、限速标识等显示装置上显示的标识，进而根据识别结果控制智能汽车的行驶。
但是，由于显示装置是周期性的点亮或熄灭，显示装置和图像采集设备是两个不同的设备，显示装置被点亮的触发时间点和图像采集设备曝光的触发时间点不能实现同步控制，因此会出现图像采集设备采集图像时的曝光时间与显示装置的被点亮时间不重叠，导致图像采集设备采集的图像中，显示装置为熄灭状态。而当前显示装置的设计是给人眼识别的，考虑了人眼的视觉暂留效应，显示装置实际要表示的结果是与人眼识别的结果一致，显示装置虽然是周期性点亮或者熄灭，但是，由于亮灭周期较短，人眼识别的结果是显示装置一直处于点亮状态。因此控制器根据图像采集设备采集的图像进行识别的识别结果与显示装置实际表示的结果会出现不一致，若MDC的控制器根据采集的图像的识别结果控制智能汽车的行驶，则可能导致智能汽车违反交通规则甚至出现安全事故。
为了解决上述问题，本申请提供一种智能汽车的控制方法，在智能汽车行驶的过程中，通过动态调整图像采集设备的采集参数，有效识别显示装置上显示的信号标识，其中，上述采集参数包括图像采集设备采集一张图像的采集周期、曝光次数、曝光时长与曝光间隔时长。在智能汽车上的图像采集设备能够采集显示装置上的信号标识图像时，通过提高图像采集设备采集一张图像的曝光次数或者提高图像采集设备采集图像的采集频率与采集一张图像的曝光次数。缩短图像采集设备相邻两次曝光之间的时间间隔，提高图像采集设备曝光时间与信号标识被点亮时间之间的重叠比例，进而提高图像采集设备采集到的图像中信号标识为点亮状态的比例，提高控制器识别的结果与信号标识实际表示结果之间相同的概率，进而提高智能汽车的安全性。在智能汽车的图像采集设备不能采集到显示装置的图像时，将图像采集设备的采集参数调整至图像采集设备不能采集显示装置上的信号标识图像时，图像采集设备的采集参数。
下面以显示装置为LED显示装置，信号标识为交通灯为例，对本申请提供的智能汽车控制方法进行详细介绍。如图1所示，图1是本申请提供的一种智能汽车的系统架构图，智能汽车10包括控制器110、图像采集设备120、通信盒子130与定位模块140，其中，图像采集设备120、通信盒子130以及定位模块140均与控制器110通信连接，控制器通过通信盒子130以及网络与云服务数据中心或者其他设备通信。如图2所示，图2是本申请提供的一种智能汽车的控制方法的流程示意图，该方法以下S201至S205。
S201、控制器110识别交通灯。
作为一种可能的实现方式，控制器110识别交通灯是指控制器在图像采集设备120采集的图像中识别是否存在交通灯。智能汽车10在行驶的过程中，控制器110在图像采集设备采集的图像中没有识别到交通灯时，图像采集设备以第一采集频率采集图像并将采集到的图像 发送给控制器110，控制器110会对接收到的图像进行交通灯检测，以确定图像中是否存在交通灯，若存在，则控制器110执行下述S202；若不存在，控制器110继续接收图像采集设备120发送的图像并进行交通灯检测。
作为另一种可能的实现方式，控制器110识别交通灯是指识别交通灯与智能汽车10之间的距离，进而根据距离确定图像采集设备120采集的图像中是否存在交通灯。控制器110可以根据定位模块140提供的智能汽车的位置信息结合地图数据，确定图像采集设备120采集到的图像中是否能够检测到交通灯。具体的，上述地图数据中包括交通灯的位置信息，控制器110获取定位模块140提供的当前智能汽车的位置信息以及地图数据中行驶路径上的交通灯的位置信息，确定智能汽车与行驶路径上最近的交通灯之间的距离。当确定智能汽车与行驶路径上最近的交通灯的距离小于或者等于预设距离时，控制器110确定图像采集设备120采集的图像中可以检测到交通灯。上述预设距离可以是根据图像采集设备120采集的图像清楚的识别出目标物体时，图像采集设备120与目标物体的距离，例如50米，100米或者200米，本申请实施例不做具体限制。
需要说明的是，控制器110在没有检测到交通灯，图像采集设备120以第一采集频率采集图像时，图像采集设备120采集一张图像的采集周期为第一采集周期T1，图像采集设备120通过一次曝光采集一张图像，即图像采集设备120的第一曝光次数为1，图像采集设备120采集一张图像的曝光时长为第一曝光时长t 1b，曝光间隔时长为第一曝光间隔时长t 1g。例如，图像采集设备120采集图像的第一采集频率为30，则第一采集周期为33.33ms，其中第一曝光时长可以为10ms，第一曝光间隔时长为23.33ms。
S202、控制器110生成调整指令并将调整指令发送给图像采集设备120。
控制器110在图像采集设备120可以采集到交通灯的图像之后，获取显示交通灯的显示装置的显示参数，该显示参数包括显示装置的亮灭周期以及每个亮灭周期内交通灯点亮的点亮时长。控制器110根据显示装置的显示参数计算得到采集参数，根据计算得到的采集参数生成调整指令并将调整指令发送给图像采集设备120。该调整指令指示图像采集设备120将当前各个采集参数的值调整为调整指令中各个采集参数对应的值，其中，采集参数包括采集周期、曝光次数、曝光时长以及曝光间隔时长，上述调整指令包括图像采集设备120的第二采集周期T2、第二曝光次数n、第二曝光时长t 2b以及第二曝光间隔时长t 2g。曝光次数指示图像采集设备120采集一张图像的曝光次数，上述第二曝光次数n为大于1的正整数，第二曝光次数n指示图像采集设备120在根据调整指令调整采集参数之后，采集一张图像需要的曝光次数。第二曝光时长t 2b表示图像采集设备120根据调整指令调整采集参数后，采集一张图像需要多次曝光时，每次曝光的曝光时长。调整指令指示图像采集设备120根据第二曝光次数n将一个采集周期内的单次曝光调整为n次曝光，根据第二曝光时长将每次曝光的时长从第一曝光时长调整至第二曝光时长，根据第二曝光间隔时长将曝光间隔时长从第一曝光间隔时长调整至第二曝光间隔时长。需要说明的是，图像采集设备120在接收到调整指令时，图像采集设备120通过单次曝光得到一张图像，单次曝光的时长为第一曝光时长。图像采集设备120在接收到调整指令，将单次的曝光时长由第一曝光时长调整至第二曝光时长后，以第二曝光时长作为图像采集设备120的曝光时长，每次曝光得到一组曝光数据，通过将n次以第二曝光时长作为曝光时长时得到的数据进行融合，得到一张图像。也就是说，图像采集设备120在根据调整指令调整曝光时长以及曝光间隔时长之后，图像采集设备120将连续多次曝光得到的数据进行融合后才能得到一张图像。其中，数据的融合是将多个次曝光获得的 数据进行合并，最终获得一张图像。具体实施时，可以采用传统技术中数据融合的方法，本申请对此不做限定。
控制器110根据显示装置的显示参数计算采集参数时，首先根据获取的显示装置的亮灭周期，确定将图像采集设备120的第二采集周期T2调整为与显示装置的亮灭周期相同。若控制器110将图像采集设备120在一个第二采集周期内总的曝光时长设置为T 2b，则一个第二采集周期内总的曝光间隔时长为T2-T 2b。为保证第二曝光间隔时长t 2g小于一个亮灭周期内交通灯点亮的时长T 亮，一个采集周期内曝光次数n满足(T2-T 2b)/n<T 亮，即n>(T2-T 2b)/T 亮，图像采集设备120的第二曝光时长t 2b＝T 2b/n，第二曝光间隔时长t 2g＝(T2-T 2b)/n。可以理解，为了防止曝光次数n取值过大，可以设置n≤T2/T 亮，即(T2-T 2b)/T 亮<n≤T2/T 亮。上述第二采集周期T2内总的曝光时长T 2b可以与第一曝光时长t 1b相同，也可以比第一曝光时长t 1b长，还可以比第一曝光时长t 1b短，本申请实施例不做具体限制。
示例性的，如图3所示，若显示装置亮灭一次的亮灭周期为20毫秒，一个亮灭周期内交通灯点亮的时长为2ms，熄灭的时长为18ms。图像采集设备在接收到调整指令前，图像采集设备120的第一采集周期为33.33ms，图像采集设备120的第一曝光时长为10ms，第一曝光间隔时长为23.33ms。在图像采集设备120以第一采集周期采集交通灯图像时，在部分采集周期内，图像采集设备120的曝光时间与交通灯灯亮的时间会有重叠，则图像中交通灯为点亮状态。而在部分采集周期内，图像采集设备120的曝光时间与交通灯灯亮的时间不会有重叠，导致图像采集设备120采集的图像中交通灯为熄灭状态。例如，图3中第N张与第N+1张图像对应的采集周期中图像采集设备120的曝光时间与交通灯灯亮的时间有重叠，则第N张与第N+1张图像中交通灯为点亮状态。而第N+2张图像对应的采集周期中图像采集设备120的曝光时间与交通灯灯亮的时间没有重叠，则第N+2张图像中交通灯为熄灭状态。其中，N为大于或等于1的正整数。
控制器110在确定可以采集到交通灯图像后，将图像采集设备120的第一采集周期33.33ms调整为第二采集周期为20ms。控制器110将图像采集设备120一个采集周期内总的曝光时长调整为12.5ms，第二曝光间隔时长为7.5ms。为了使曝光间隔时长小于2ms，控制器110将12.25ms的曝光时长切分为5次，则图像采集设备120的每个第二曝光时长为2.5ms，曝光间隔时长为1.5ms。图像采集设备120每次打开快门2.5ms完成一次曝光，然后间隔1.5ms再次将快门打开2.5ms，从图像采集设备120将第一曝光时长调整为第二曝光时长之后，当第5次打开快门并关闭之后，将之前5次的曝光得到的数据进行融合，得到一张图像，并将该图像发送给控制器110。图像采集设备120在第5次打开快门并关闭之后，再隔1.5ms再次打开快门。即图像采集设备120在第5m次打开快门并关闭之后，将第(5m-4)次至第5m次之间5次曝光得到的数据进行融合得到一张图像并发送给控制器110，其中，m为正整数。如图3所示，由于第二曝光间隔时长小于交通灯在一个亮灭周期内灯亮的时长，即使交通灯被点亮的时间点与曝光结束的时间点为同一时刻，由于交通灯亮的时长比曝光间隔时长要长，图像采集设备120在下一次曝光时，交通灯依然是亮着的，因此可以保证一个采集周期内曝光时长与交通灯亮的时长具有重叠部分，在上述举例中，图像采集周期的曝光时长与交通灯亮的时长的重叠时间最少为0.5ms，最长为2ms。需要说明的是，上述示例仅用作举例，不能理解为具体限定。
根据上述方法，将图像采集设备120的第二采集周期调整为等于显示装置的亮灭周期，并使第二曝光间隔时长小于显示装置一个亮灭周期内点亮的时长。可以确保图像采集设备120 在一个采集周期内的n次曝光中至少有一次的曝光时间与交通灯点亮的时间有重叠，确保融合n次曝光的数据得到的图像中交通灯为灯亮状态，即可以确保图像采集设备120输出的每张图像中交通灯均为灯亮状态。
上述控制器110获取显示交通灯的显示装置的显示参数的方法包括但不限于以下三种方式：第一种方式是显示装置中配置有发送模块，该发送模块周期性的将其亮灭周期以及一个亮灭周期内点亮的时长进行广播，接近该显示装置的智能汽车均能接收到上述发送模块发送的数据，从而获取到上述显示装置的显示参数。第二种方式是控制器110在确定可以采集到交通灯的图像后，通过通信盒子130向显示装置发送获取请求，该获取请求用于获取上述显示参数，显示装置在接收到该获取请求之后，将显示参数发送给通信盒子130。第三种方式是控制器110在确定可以采集到交通灯的图像后，控制器110通过通信盒子130向云服务数据中心发送获取请求，该获取请求用于获取上述显示参数，该获取请求包括显示装置的位置信息，云服务数据中心在接收到该获取请求之后，根据获取请求中的位置信息以及云服务数据中心中存储的位置信息与显示装置的对应关系，确定获取请求中的位置关系对应的显示装置，然后获取该位置信息对应的显示装置的显示参数并将显示参数发送给控制器110。
S203、图像采集设备120接收调整指令，根据调整指令调整采集参数，根据调整后的采集参数采集图像并将采集到的图像发送给控制器110。
图像采集设备120在接收到控制器110发送的调整指令之后，调整图像采集设备120的采集周期，曝光次数、曝光时长以及曝光间隔时长，以调整后的采集参数采集图像。图像采集设备120在一个采集周期内进行多次曝光，然后将多次曝光得到的数据进行融合，得到该采集周期内采集的一张图像，并将该图像发送给控制器110。可以理解，为了使控制器110能够根据图像识别出交通灯是红灯、黄灯或者绿灯，图像采集设备120采集的图像为彩色图像，例如红绿蓝(red green blue，RGB)色彩模式的图像。
S204、控制器110接收图像采集设备120发送的图像，对图像中的交通灯进行识别，确定交通灯的状态。
控制器110在接收到图像采集设备120采集的图像之后，首先确定图像中交通灯的位置信息，上述位置信息为图像中四个像素点的坐标值，例如，这四个像素点的坐标值分别为(u1，v1)、(u1，v2)、(u2，v1)以及(u2，v2)，这四个像素点确定的一个初始矩形区域内包括交通灯。控制器110根据上述交通灯在图像中的位置信息确定图像中包括交通灯的感兴趣区域，例如将上述初始矩形区域作为感兴趣区域，或者根据上述四个像素点坐标确定四个新的像素点坐标，将上述初始矩形区域扩大预设倍数之后得到的新的矩形区域作为感兴趣区域，以防止初始矩形区域没有包含完整的交通灯。例如，四个新的像素点坐标分别为(u1-1/2(u2-u1)，v1-1/2(v2-v1))、(u1-1/2(u2-u1)，v1+1/2(v2-v1))、(u1+1/2(u2-u1)，v1-1/2(v2-v1))以及(u1+1/2(u2-u1)，v1+1/2(v2-v1))，将初始矩形区域扩大为4倍之后作为感兴趣区域，然后截取上述感兴趣区域作为目标图像，将目标图像输入至交通灯识别网络中，对目标图像中的交通灯进行识别，得到识别结果。
控制器110根据上述目标图像进行识别的识别结果包括红灯、绿灯、黄灯以及未知状态，未知状态表示图像中没有交通灯或者图像中交通灯处于熄灭状态。而控制器110需要确定交通灯的状态，即交通灯是红灯、绿灯还是黄灯，进而根据交通灯的状态确定智能汽车是否需要停车或者减速行驶。因此，当控制器110根据图像识别得到当前一张图像的识别结果之后，需要根据该张图像之前的交通灯状态确定该张图像中交通灯的状态，将交通灯状态确定为红 灯、绿灯或者黄灯中的任意一种。其中，控制器可以通过目标图像中交通灯的颜色信息对交通灯进行识别，也可以通过训练好的用于识别交通灯的神经网络模型对目标图像中的交通灯进行识别，例如卷积神经网络模型(convolutional neural networks，CNN)、循环神经网络模型(recurrent neural network，RNN)等。
具体的，若控制器110对当前一张图像的识别结果为红灯、绿灯以及黄灯中的任意一种，则控制器110以该张图像的识别结果为交通灯的状态。若控制器110对当前一张图像的识别结果是未知状态，则将该张图像之前确定的交通灯的状态作为当前一张图像的对应的交通灯的状态，例如，当前一张图像的识别结果为未知状态，在该张图像之前交通灯的状态为绿灯，则将绿灯作为当前一张图像的对应的交通灯的状态。控制器110根据当前一张图像的识别结果以及该张图像之前的交通灯的状态确定该张图像中交通灯的状态的方法如下表1所示，其中，表中的任意状态表示红灯、绿灯、黄灯或未知状态中的任意一种状态。
表1
当前一张之前交通灯的状态 当前一张识别结果 交通灯的状态
任意状态 绿灯 绿灯
任意状态 黄灯 黄灯
任意状态 红灯 红灯
绿灯 未知状态 绿灯
黄灯 未知状态 黄灯
红灯 未知状态 红灯
作为一种可能的实现方式，上述S204中，控制器110在接收到图像采集设备120发送的图像之后，可以通过检测图像中的交通灯以确定交通灯在图像中的位置信息，例如，控制器110可以根据颜色信息和/或形状信息确定图像中交通灯的位置信息。
作为另一种可能的实现方式，控制器110还可以通过上述定位模块140以及图像数据确定交通灯相对于智能汽车的位置信息，根据交通灯相对于智能汽车的位置信息确定交通灯相对于车辆坐标系的坐标，然后结合车辆坐标系与像素坐标系之间的转换关系确定交通灯在图像中的位置信息。其中，上述交通灯相对于智能汽车的位置信息包括交通灯相对于车辆的距离、交通灯的高等、交通灯的长度以及车辆距离道路边缘的距离，其中，交通灯的高度是指交通灯的下侧边缘与路面的距离，交通灯的长度是指交通灯左侧(相对于车辆)距离道路边缘的距离。具体的，如图4所示，图4是本申请提供的一种车辆坐标系的示意图，在图4所示的车身俯视图中，车辆坐标系以车辆车后轴为Y w轴，以车后轴与车辆车轴中心线的交点为坐标原点，车辆的车轴中心线作为X w轴，以车辆行驶方向为X w轴的正方向，其中，X wO wY w组成的平面与水平面平行，以垂直于平面X wO wY w的方向作为Z w轴方向，以垂直平面X wO wY w向上的方向为Z w轴正方向。
根据上述交通灯相对于智能汽车的位置信息可以确定交通灯相对于车辆坐标系的位置，示例性的，如图5所示，图5是本申请提供的一种车辆坐标系与交通灯的位置关系示意图。交通灯相对于车辆的距离为交通灯在X w轴上的坐标值x，交通灯的高度为交通灯在Z w轴上的坐标值z，车辆距离道路边缘的距离为交通灯右侧(相对于车辆)在Y w轴上的坐标值y1，交通灯的长度与车辆距离道路边缘的距离的差值为交通灯左侧(相对于车辆)在Y w轴上的坐标值y2。由于交通灯本身具有一定的宽度L，则包括交通灯的矩形区域的四个顶点在车辆坐 标系中的坐标分别为A(x，y1，z)、B(x，y1，z+L)、C(x，y2，z)与D(x，y2，z+L)，上述四个坐标点的坐标即为交通灯在三维坐标系中的坐标，根据车辆坐标系与像素坐标系之间的转换关系，即可确定上述四个坐标点在像素坐标系中的坐标，即为上述交通灯在图像中的位置信息。
S205、控制器110根据交通灯的状态控制智能汽车的行驶。
控制器110在识别出交通灯的状态后，根据交通灯的状态控制智能汽车的行驶。示例性的，在交通灯状态为绿灯时，控制智能车保持当前速度行驶；在交通灯为黄灯或者红灯时，控制智能车逐渐减速至停车。
可以理解，控制器110在根据上述S201～S205中的方法识别交通灯的状态，并根据交通灯的状态控制智能汽车的行驶后，控制器110在图像采集设备120采集的图像中检测不到显示装置时，生成回复指令并将回复指令发送给图像采集设备120，使图像采集设备将采集参数调整至接收到上述调整指令之前的数值。
作为一种可能的实现方式，控制器110在采用上述S201中的方法识别到交通灯后，控制器110不能获取到显示装置的亮灭周期以及显示装置在一个亮灭周期内灯亮的时长。控制器110将显示装置通常采用的亮灭周期作为当前识别到的交通灯的显示装置的亮灭周期，将显示装置通常采用的一个亮灭周期内灯亮的时长，作为当前交通灯在一个亮灭周期内灯亮的时长。例如，显示装置通常的亮灭周期为20ms，一个亮灭周期内灯亮的时长为2ms，则认为当前交通灯的亮灭周期设置为20ms，一个亮灭周期内交通灯灯亮的时长为2ms。然后，控制器110按照上述S202中确定第二采集周期、第二曝光次数、第二曝光时长以及第二曝光间隔时长相同的方法，确定图像采集设备120调整后的采集参数。
作为一种可能的实现方式，控制器110在采用上述S201中的方法识别到交通灯后，控制器110不能获取到显示装置的亮灭周期以及显示装置在一个亮灭周期内灯亮的时长。控制器110获取图像采集设备120当前的第一采集周期，若第一采集周期小于或者等于预设采集周期，也就是说，图像采集设备120在单位时间内采集的图像的数量大于或者等于预设数量。控制器110确定图像采集设备120保持第一采集周期不变。若第一采集周期大于预设采集周期，也就是说，图像采集设备120在单位时间内采集的图像的数量小于预设数量，控制器110确定将图像采集设备120的采集周期由第一采集周期降低到预设采集周期。其中，预设采集周期为显示装置通常采用的亮灭周期，例如，显示装置通常的亮灭周期为20ms，则预设采集周期取值为20ms。同时，控制器110将显示装置通常采用的一个亮灭周期内灯亮的时长，作为当前交通灯在一个亮灭周期内灯亮的时长。控制器110根据上述方法确定图像采集设备120调整采集参数后的采集周期以及显示装置在一个亮灭周期内灯亮的时长，根据上述S202中确定第二采集周期、第二曝光次数、第二曝光时长以及第二曝光间隔时长相同的方法，确定图像采集设备120调整后的采集参数，其中，第一采集周期小于或者等于预设采集周期时的第一采集周期，或者第一采集周期大于预设采集周期时的第二采集周期，即为上述S202中计算时的第二采集周期T2。预设采集周期可以根据历史数据的统计信息经过计算获得，也可以是经验值。
作为一种可能的实现方式，控制器110在采用上述S201中的方法识别到交通灯后，控制器110无论能否获取到显示装置的亮灭周期以及显示装置在一个亮灭周期内点亮的时长，控制器110生成的调整指令中的采集周期、曝光次数、曝光时长以及曝光间隔时长均为预设的固定值，例如，采集周期为20ms、曝光次数为5次、曝光时长为2.5ms，曝光间隔时长为1.5ms。 图像采集设备120在接收到调整指令后，将采集参数调整为上述预设的固定值。
值得说明的是，对于上述方法实施例，为了简单描述，故将其都表述为一系列的动作组合，但是本领域技术人员应该知悉，本发明并不受所描述的动作顺序的限制，其次，本领域技术人员也应该知悉，说明书中所描述的实施例均属于优选实施例，所涉及的动作并不一定是本发明所必须的。
本领域的技术人员根据以上描述的内容，能够想到的其他合理的步骤组合，也属于本发明的保护范围内。其次，本领域技术人员也应该熟悉，说明书中所描述的实施例均属于优选实施例，所涉及的动作并不一定是本发明所必须的。
上文中结合图1至图5详细描述了根据本申请所提供的智能汽车的控制方法，下面将结合图6至图7，描述本申请所提供的装置及相关设备。
图6是本申请提供的一种智能汽车的控制装置的结构示意图，该控制装置用于包括图像采集设备与定位模块的智能汽车中，其中，图像采集设备安装于车辆上，用于采集车辆行驶方向上的图像，该控制装置与图像采集设备连接，用于接收图像采集装置采集的图像。该控制装置600包括：处理单元610、发送单元620以及获取单元630，其中，
处理单元610，用于在识别信号标识的显示装置时，计算智能汽车中图像采集设备的采集参数的目标值；图像采集设备的采集参数用于指示图像采集设备采集图像的采集周期、曝光次数、曝光时长与曝光间隔时长。处理单元610可以通过获取单元630获取图像采集设备采集的图像，根据图像采集设备采集的图像识别信号标识的显示装置，也可以根据定位模块提供的位置信息与信号标识的距离识别信号标识的显示装置。具体的，处理单元610识别信号标识的显示装置可以参照上述S201中控制器110识别上述显示装置的相关描述。
处理单元610在在识别到信号标识的显示装置时，确定需要对图像采集设备的采集参数进行调整，图像采集设备通过获取单元630获取显示装置的亮灭周期以及显示装置显示的信号标识在一个亮灭周期内点亮的时长，亮灭周期包括信号标识点亮的时长与信号标识熄灭的时长。然后根据显示装置的亮灭周期以及显示装置显示的信号标识在一个亮灭周期内点亮的时长，计算图像采集设备的采集周期、曝光次数、曝光时长与曝光间隔时长等采集参数的目标值，具体的，处理单元610根据计算采集参数的方法可以参照上述S202中控制器110计算采集参数的相关描述，在此不再赘述。
处理单元610根据采集参数的目标值，生成调整指令，该调整指令指示图像采集设备根据调整指令中采集参数的目标值调整采集参数。
发送单元620，用于向图像采集设备发送调整指令。
应理解的是，本申请实施例的控制装置600可以通过专用集成电路(application-specific integrated circuit，ASIC)实现，或可编程逻辑器件(programmable logic device，PLD)实现，上述PLD可以是复杂程序逻辑器件(complex programmable logical device，CPLD)，现场可编程门阵列(field-programmable gate array，FPGA)，通用阵列逻辑(generic array logic，GAL)或其任意组合。也可以通过软件实现图2所示的控制方法时，控制装置600及其各个模块也可以为软件模块。
可选地，处理单元610，还用于根据图像采集设备调整采集参数后获得的图像识别信号标识，并根据识别的信号标识控制智能车行驶。处理单元610根据图像采集设备采集的图像对信号标识的识别方法可参照上述S204中对交通灯的识别方法，在此不再赘述。控制装置600在识别出交通灯的状态后，根据交通灯的状态控制智能汽车的行驶。示例性的，在交通 灯状态为绿灯时，控制智能车保持当前速度行驶；在交通灯为黄灯或者红灯时，控制智能车逐渐减速至停车。
具体的，上述控制装置600所执行的操作可参照上述方法实施例中控制器100识别信号标识、计算采集参数以及识别交通灯等相关操作，在此不再具体描述。
图7为本申请提供的一种计算设备的结构示意图，该计算设备700至少包括：处理器710、通信接口720以及存储器730，处理器710、通信接口720以及存储器730通过总线740相互连接，其中，该处理器710用于执行该存储器730存储的指令。该存储器730存储程序代码，且处理器710可以调用存储器720中存储的程序代码执行以下操作：
当控制器识别信号标识的显示装置时，计算所述智能汽车中图像采集设备的采集参数的目标值；所述图像采集设备的采集参数用于指示所述图像采集设备采集图像的采集周期、曝光次数、曝光时长与曝光间隔时长；
向所述图像采集设备发送调整指令，所述调整指令指示所述图像采集设备根据所述调整指令中采集参数的目标值调整采集参数；
根据所述图像采集设备调整采集参数后获得的图像识别信号标识，并根据识别的信号标识控制智能车行驶。
应理解，在本申请实施例中处理器710可以有多种具体实现形式，例如处理器710可以为中央处理器(central processing unit，CPU)或图像处理器(graphics processing unit，GPU)，处理器710还可以是单核处理器或多核处理器。处理器710可以由CPU和硬件芯片的组合。上述硬件芯片可以是专用集成电路(application-specific integrated circuit，ASIC)，可编程逻辑器件(PLD)或其组合。上述PLD可以是复杂可编程逻辑器件(CPLD)，现场可编程逻辑门阵列(FPGA)，通用阵列逻辑(GAL)或其任意组合。处理器710也可以单独采用内置处理逻辑的逻辑器件来实现，例如FPGA或数字信号处理器(digital signal processor，DSP)等。
通信接口720可以为有线接口或无线接口，用于与其他模块或设备进行通信，有线接口可以是以太接口、控制器局域网络(controller area network，CAN)接口、局域互联网络(local interconnect network，LIN)以及FlexRay接口，无线接口可以是蜂窝网络接口或使用无线局域网接口等。例如，本申请实施例中通信接口720具体可用于接收图像采集设备发送的图像数据、接收显示装置发送的亮灭周期等。
存储器730可以是非易失性存储器，例如，只读存储器(read-only memory，ROM)、可编程只读存储器(programmable ROM，PROM)、可擦除可编程只读存储器(erasable PROM，EPROM)、电可擦除可编程只读存储器(electrically EPROM，EEPROM)或闪存。存储器730也可以是易失性存储器，易失性存储器可以是随机存取存储器(random access memory，RAM)，其用作外部高速缓存。
存储器730也可用于存储指令和数据，以便于处理器710调用存储器730中存储的指令实现上述处理单元610执行的操作，例如上述方法实施例中控制器110执行的操作。此外，计算设700可能包含相比于图7展示的更多或者更少的组件，或者有不同的组件配置方式。
总线740可以是CAN总线或其他实现车内各个系统或设备之间互连的内部总线。总线740可以分为地址总线、数据总线、控制总线等。为便于表示，图7中仅用一条粗线表示，但并不表示仅有一根总线或一种类型的总线。
可选地，该计算设备700还可以包括输入/输出接口750，输入/输出接口750连接有输入/输出设备，用于接收输入的信息，输出操作结果。
应理解，根据本申请实施例的计算设备700可对应于本申请实施例中的控制装置600，并可以对应于执行图2所示的方法中的相应操作，并且控制装置600中的各个模块的操作和/或功能分别为了实现图1至图5中的各个方法的相应流程，为了简洁，在此不再赘述。
本申请实施例还提供一种非瞬态计算机存储介质，计算机存储介质中存储有指令，当其在处理器上运行时，可以实现上述方法实施例中的方法步骤，计算机存储介质的处理器在执行上述方法步骤的具体实现可参照上述方法实施例的具体操作，在此不再赘述。
在上述实施例中，对各个实施例的描述都各有侧重，某个实施例中没有详述的部分，可以参见其它实施例的相关描述。
上述实施例，可以全部或部分地通过软件、硬件、固件或其他任意组合来实现。当使用软件实现时，上述实施例可以全部或部分地以计算机程序产品的形式实现。所述计算机程序产品包括一个或多个计算机指令。在计算机上加载或执行所述计算机程序指令时，全部或部分地产生按照本申请实施例所述的流程或功能。所述计算机可以为通用计算机、专用计算机、计算机网络、或者其他可编程装置。所述计算机指令可以存储在计算机可读存储介质中，或者从一个计算机可读存储介质向另一个计算机可读存储介质传输，例如，所述计算机指令可以从一个网站站点、计算机、服务器或数据中心通过有线(例如同轴电缆、光纤、数字用户线)或无线(例如红外、无线、微波等)方式向另一个网站站点、计算机、服务器或数据中心进行传输。所述计算机可读存储介质可以是计算机能够存取的任何可用介质或者是包含一个或多个可用介质集合的服务器、数据中心等数据存储设备。所述可用介质可以是磁性介质(例如，软盘、硬盘、磁带)、光介质、或者半导体介质。半导体介质可以是固态硬盘(solid state drive，SSD)。
以上所述，仅为本申请的具体实施方式。熟悉本技术领域的技术人员根据本申请提供的具体实施方式，可想到变化或替换，都应涵盖在本申请的保护范围之内。

Claims (10)

  1. 一种智能汽车的控制方法，其特征在于，所述方法包括：
    当控制器识别信号标识的显示装置时，所述控制器计算所述智能汽车中图像采集设备的采集参数的目标值；所述图像采集设备的采集参数用于指示所述图像采集设备采集图像的采集周期、曝光次数、曝光时长与曝光间隔时长；
    所述控制器向所述图像采集设备发送调整指令，所述调整指令指示所述图像采集设备根据所述调整指令中采集参数的目标值调整采集参数；
    所述控制器根据所述图像采集设备调整采集参数后获得的图像识别信号标识，并根据识别的信号标识控制智能车行驶。
  2. 根据权利要求1所述的方法，其特征在于，所述当控制器识别信号标识的显示装置时，所述控制器计算所述智能汽车中图像采集设备的采集参数的目标值，包括：
    所述控制器获取所述图像采集设备调整采集参数前获得的图像，对所述图像采集设备调整采集参数前获得的图像进行信号标识检测，当检测到信号标识时，所述控制器计算所述智能汽车中图像采集设备的采集参数的目标值；或者，
    所述控制器获取所述智能汽车与所述显示装置的距离，当所述距离小于或者等于预设距离时，所述控制器计算所述智能汽车中图像采集设备的采集参数的目标值，其中，所述显示装置位于所述智能汽车的前进方向上。
  3. 根据权利要求1或2所述的方法，其特征在于，所述控制器计算所述智能汽车中图像采集设备的采集参数的目标值，包括：
    所述控制器获取所述显示装置的亮灭周期以及所述信号标识在一个亮灭周期内点亮的时长，所述亮灭周期包括信号标识点亮的时长与信号标识熄灭的时长；
    所述控制器将所述亮灭周期确定为所述图像采集设备调整采集参数后的采集周期，所述采集周期为所述图像采集设备采集一张图像所用的时长；
    所述控制器根据所述图像采集设备调整采集参数后的采集周期、所述图像采集设备在一个采集周期内总的曝光时长以及所述信号标识在一个亮灭周期内点亮的时长，确定图像采集设备调整采集参数后，一个采集周期内的曝光次数；
    所述控制器根据所述图像采集设备调整采集参数后在一个采集周期内总的曝光时长以及一个采集周期内的曝光次数，确定所述图像采集设备调整采集参数后每次的曝光时长；并根据所述图像采集设备调整采集参数后的采集周期、总的曝光时长以及曝光次数，确定所述图像采集设备调整采集参数后每次的曝光间隔时长；
    所述控制器根据所述图像采集设备调整采集参数后的采集周期、每次的曝光时长、一个采集周期内的曝光次数以及每次的曝光间隔时长生成所述调整指令。
  4. 根据权利要求1至3任一项所述的方法，其特征在于，所述信号标识为交通灯，所述控制器根据所述图像采集设备调整采集参数后获得的图像识别信号标识，并根据识别的信号标识控制智能车行驶，包括：
    所述控制器根据所述图像采集设备调整采集参数后获得的图像对交通灯进行识别，得到 识别结果，所述识别结果包括红灯、绿灯、黄灯和未知状态，每张图像对应的识别结果为红灯、绿灯、黄灯或未知状态中的任意一种；
    所述控制器根据所述识别结果以及上一次确定的交通灯的状态，确定所述图像采集设备调整采集参数后获得的图像中交通灯的状态，所述交通灯的状态包括红灯、绿灯和黄灯，每张图像对应的交通灯的状态为红灯、绿灯或黄灯中的任意一种。
  5. 一种智能汽车的控制装置，其特征在于，所述装置包括：
    处理单元，用于：在识别信号标识的显示装置时，计算所述智能汽车中图像采集设备的采集参数的目标值；所述图像采集设备的采集参数用于指示所述图像采集设备采集图像的采集周期、曝光次数、曝光时长与曝光间隔时长；
    根据所述采集参数的目标值，生成调整指令，所述调整指令指示所述图像采集设备根据所述调整指令中采集参数的目标值调整采集参数；
    发送单元，用于向所述图像采集设备发送所述调整指令；
    所述处理单元，还用于根据所述图像采集设备调整采集参数后获得的图像识别信号标识，并根据识别的信号标识控制智能车行驶。
  6. 根据权利要求5所述的装置，其特征在于，所述装置还包括：
    获取单元，用于获取所述图像采集设备调整采集参数前获得的图像；
    所述处理单元，具体用于对所述图像采集设备调整采集参数前获得的图像进行信号标识检测，当检测到信号标识时，计算所述智能汽车中图像采集设备的采集参数的目标值；或者，
    所述获取单元，用于获取所述智能汽车与所述显示装置的距离；
    所述处理单元，具体用于当所述距离小于或者等于预设距离时，所述控制器计算所述智能汽车中图像采集设备的采集参数的目标值，其中，所述显示装置位于所述智能汽车的前进方向上。
  7. 根据权利要求5或6所述的装置，其特征在于，
    所述获取单元，还用于获取所述显示装置的亮灭周期以及所述信号标识在一个亮灭周期内点亮的时长，所述亮灭周期包括信号标识点亮的时长与信号标识熄灭的时长；
    所述处理单元，具体用于：将所述亮灭周期确定为所述图像采集设备调整采集参数后的采集周期，所述采集周期为所述图像采集设备采集一张图像所用的时长；
    根据所述图像采集设备调整采集参数后的采集周期、所述图像采集设备在一个采集周期内总的曝光时长以及所述信号标识在一个亮灭周期内点亮的时长，确定图像采集设备调整采集参数后，一个采集周期内的曝光次数；
    根据所述图像采集设备调整采集参数后在一个采集周期内总的曝光时长以及一个采集周期内的曝光次数，确定所述图像采集设备调整采集参数后每次的曝光时长；并根据所述图像采集设备调整采集参数后的采集周期、总的曝光时长以及曝光次数，确定所述图像采集设备调整采集参数后每次的曝光间隔时长；
    根据所述图像采集设备调整采集参数后的采集周期、每次的曝光时长、一个采集周期内的曝光次数以及每次的曝光间隔时长生成所述调整指令。
  8. 根据权利要求5-7任一项所述的装置，其特征在于，所述信号标识为交通灯，所述处理单元具体用于：
    根据所述图像采集设备调整采集参数后获得的图像对交通灯进行识别，得到识别结果，所述识别结果包括红灯、绿灯、黄灯和未知状态，每张图像对应的识别结果为红灯、绿灯、黄灯或未知状态中的任意一种；
    根据所述识别结果以及上一次确定的交通灯的状态，确定所述图像采集设备调整采集参数后获得的图像中交通灯的状态，所述交通灯的状态包括红灯、绿灯和黄灯，每张图像对应的交通灯的状态为红灯、绿灯或黄灯中的任意一种。
  9. 一种计算设备，其特征在于，包括处理器和存储器；所述存储器用于存储指令，所述处理器用于执行所述指令，当所述处理器执行所述指令时，所述处理器执行如权利要求1至4任一项所述的方法。
  10. 一种非瞬态的计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，所述计算机程序被计算设备执行时实现如权利要求1至4任一项所述的方法。
PCT/CN2021/073282 2020-01-23 2021-01-22 智能汽车的控制方法、装置及相关设备 Ceased WO2021147995A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
EP21744507.1A EP4086874A4 (en) 2020-01-23 2021-01-22 Control method and apparatus for intelligent vehicle, and related device

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010077052.4A CN113160590B (zh) 2020-01-23 2020-01-23 智能汽车的控制方法、装置及相关设备
CN202010077052.4 2020-01-23

Publications (1)

Publication Number Publication Date
WO2021147995A1 true WO2021147995A1 (zh) 2021-07-29

Family

ID=76882153

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/073282 Ceased WO2021147995A1 (zh) 2020-01-23 2021-01-22 智能汽车的控制方法、装置及相关设备

Country Status (3)

Country Link
EP (1) EP4086874A4 (zh)
CN (1) CN113160590B (zh)
WO (1) WO2021147995A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115914097B (zh) * 2021-08-11 2025-01-07 深圳智岩科技股份有限公司 控制发光设备的方法及相关设备
CN113627362B (zh) * 2021-08-13 2025-02-18 深圳市道通科技股份有限公司 指示信号识别方法、装置、设备以及计算机存储介质
JP7317081B2 (ja) * 2021-08-23 2023-07-28 本田技研工業株式会社 進行停止判定装置

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2009017293A (ja) * 2007-07-05 2009-01-22 Panasonic Corp 撮像装置
CN102196184A (zh) * 2010-03-12 2011-09-21 株式会社日立制作所 摄像装置
CN106303308A (zh) * 2015-06-25 2017-01-04 三星钻石工业股份有限公司 摄像装置
US20180253613A1 (en) * 2017-03-06 2018-09-06 Honda Motor Co., Ltd. System and method for vehicle control based on red color and green color detection
WO2019039022A1 (ja) * 2017-08-21 2019-02-28 ソニーセミコンダクタソリューションズ株式会社 撮像装置及び機器
CN111291620A (zh) * 2020-01-14 2020-06-16 北京小马智行科技有限公司 识别光源的方法、装置和系统

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102010003039A1 (de) * 2010-03-18 2011-09-22 Basler Ag Farbsättigungskorrektur
CN101977284B (zh) * 2010-11-29 2012-07-18 杭州海康威视软件有限公司 摄像系统和摄像方法
CN101977285B (zh) * 2010-11-29 2015-09-02 杭州海康威视数字技术股份有限公司 摄像系统和摄像方法
JP5680573B2 (ja) * 2012-01-18 2015-03-04 富士重工業株式会社 車両の走行環境認識装置
DE102012221159A1 (de) * 2012-11-20 2014-05-22 Robert Bosch Gmbh Verfahren und Vorrichtung zum Erkennen von Wechselverkehrszeichen
US20170041591A1 (en) * 2013-12-25 2017-02-09 Hitachi Automotive Systems ,Ltd. Vehicle-Mounted Image Recognition Device
CN105635597B (zh) * 2015-12-21 2018-07-27 湖北工业大学 车载相机的自动曝光方法及系统
US10453208B2 (en) * 2017-05-19 2019-10-22 Waymo Llc Camera systems using filters and exposure times to detect flickering illuminated objects

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2009017293A (ja) * 2007-07-05 2009-01-22 Panasonic Corp 撮像装置
CN102196184A (zh) * 2010-03-12 2011-09-21 株式会社日立制作所 摄像装置
CN106303308A (zh) * 2015-06-25 2017-01-04 三星钻石工业股份有限公司 摄像装置
US20180253613A1 (en) * 2017-03-06 2018-09-06 Honda Motor Co., Ltd. System and method for vehicle control based on red color and green color detection
WO2019039022A1 (ja) * 2017-08-21 2019-02-28 ソニーセミコンダクタソリューションズ株式会社 撮像装置及び機器
CN111291620A (zh) * 2020-01-14 2020-06-16 北京小马智行科技有限公司 识别光源的方法、装置和系统

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4086874A4

Also Published As

Publication number Publication date
EP4086874A4 (en) 2023-05-31
EP4086874A1 (en) 2022-11-09
CN113160590A (zh) 2021-07-23
CN113160590B (zh) 2023-02-03

Similar Documents

Publication Publication Date Title
US12539805B2 (en) Intelligent light switching method and system, and related device
CN110430401B (zh) 车辆盲区预警方法、预警装置、mec平台和存储介质
US12397826B2 (en) Method and apparatus for detecting drivable area, mobile device and storage medium
CN106647776B (zh) 车辆变道趋势的判断方法、判断装置和计算机存储介质
CN112149550A (zh) 一种基于多传感器融合的自动驾驶车辆3d目标检测方法
WO2021147995A1 (zh) 智能汽车的控制方法、装置及相关设备
CN106934347B (zh) 障碍物识别方法及装置、计算机设备及可读介质
CN109840454B (zh) 目标物定位方法、装置、存储介质以及设备
CN117591847B (zh) 基于车况数据的模型指向评测方法和装置
CN118429952B (zh) 用于交通管理的智能车牌识别平台
WO2022188292A1 (zh) 目标检测及控制方法、系统、设备及存储介质
CN106295494A (zh) 车外环境识别装置
WO2022246860A1 (zh) 一种自动驾驶系统的性能测试方法
CN116563801A (zh) 交通事故检测方法、装置、电子设备和介质
CN115675479A (zh) 用于利用对象分类器运行机动车中的自动驾驶功能的方法和处理器设备、及机动车
CN116091542A (zh) 一种基于多相机的车辆跟踪方法与装置
CN118762535B (zh) 基于深度学习的交通信号灯智能调控方法及系统
CN117152718B (zh) 交通灯响应方法、装置、车辆及计算机可读存储介质
KR102116029B1 (ko) 드론을 이용한 교통신호 최적화 시스템
US20250076027A1 (en) Obstacle ranging method and device, vehicle and medium
CN118470681A (zh) 交通灯检测方法、装置、电子设备及存储介质
CN109017787B (zh) 一种行驶控制方法
CN111218894B (zh) 一种发光引导设施的控制方法、装置、电子设备及系统
CN117593892B (zh) 真值数据的获取方法和装置、存储介质及电子设备
CN116434041B (zh) 错误感知数据的挖掘方法、装置、设备及自动驾驶车辆

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21744507

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2021744507

Country of ref document: EP

Effective date: 20220803

NENP Non-entry into the national phase

Ref country code: DE