WO2020177571A1 - 建立车辆纵向运动模型的方法、装置及计算机系统 - Google Patents

建立车辆纵向运动模型的方法、装置及计算机系统 Download PDF

Info

Publication number
WO2020177571A1
WO2020177571A1 PCT/CN2020/076530 CN2020076530W WO2020177571A1 WO 2020177571 A1 WO2020177571 A1 WO 2020177571A1 CN 2020076530 W CN2020076530 W CN 2020076530W WO 2020177571 A1 WO2020177571 A1 WO 2020177571A1
Authority
WO
WIPO (PCT)
Prior art keywords
vehicle
acceleration
information
angle
quantized value
Prior art date
Application number
PCT/CN2020/076530
Other languages
English (en)
French (fr)
Inventor
赵春明
Original Assignee
阿里巴巴集团控股有限公司
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 阿里巴巴集团控股有限公司 filed Critical 阿里巴巴集团控股有限公司
Publication of WO2020177571A1 publication Critical patent/WO2020177571A1/zh

Links

Images

Classifications

    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W40/00Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
    • B60W40/10Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to vehicle motion
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W40/00Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
    • B60W40/10Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to vehicle motion
    • B60W40/107Longitudinal acceleration
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W50/00Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W50/00Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B60W2050/0001Details of the control system
    • B60W2050/0019Control system elements or transfer functions
    • B60W2050/0028Mathematical models, e.g. for simulation
    • B60W2050/0031Mathematical model of the vehicle

Definitions

  • This application relates to the field of automatic driving technology, in particular to a method, device and computer system for establishing a vehicle longitudinal motion model.
  • the control of the vehicle mainly includes the control of the driving speed and direction.
  • the automatic control of vehicle speed and the distance between the vehicle and front and rear vehicles or obstacles belongs to a kind of longitudinal control.
  • Lateral control refers to control perpendicular to the direction of movement, which is steering control for automobiles.
  • the longitudinal control problem can be attributed to the control of engine output and brakes.
  • the traditional realization scheme through the combination of various engine models, car operating models and braking process models with different controller algorithms, an approximate linear model of the engine and car motion processes is established, and the controller is designed on this basis.
  • the control achieved by this method requires accurate knowledge of the parameters of multiple main modules such as the driver and brake of the vehicle. Therefore, the parameter dependence is large and the model error is large, so the accuracy is poor and the adaptability is poor.
  • the present application provides a method, a device and a computer system for establishing a longitudinal motion model of a vehicle, which can realize more accurate automatic driving control through a simpler and higher-precision longitudinal motion model.
  • An automatic driving control method including:
  • the longitudinal motion model includes the corresponding relationship between quantized value information, speed, angle, and acceleration of the acceleration/brake control component of the vehicle;
  • the longitudinal motion model determine the quantized value information of the corresponding acceleration/brake control component under the conditions of the target acceleration information, current speed, and angle information;
  • a method for establishing a vehicle longitudinal motion model including:
  • the vehicle Under the condition of the target sampling angle, the vehicle is gradually accelerated to the highest safe driving speed under the target sampling angle condition by controlling the quantized value of the acceleration control component, and then the vehicle is gradually decelerated to a stop by controlling the quantized value of the braking control component ;
  • the vehicle longitudinal motion model is established according to multiple sets of the corresponding relationships obtained under multiple target sampling angle conditions.
  • An automatic driving control method including:
  • An automatic driving control device includes:
  • the model obtaining unit is configured to obtain a longitudinal motion model of the vehicle, wherein the longitudinal motion model includes the corresponding relationship between the quantized value, speed, angle, and acceleration of the acceleration/brake control component of the vehicle;
  • the information obtaining unit is used to receive the target acceleration information output by the motion control algorithm, and obtain the current speed and angle information of the vehicle;
  • a quantized value information determining unit configured to determine, according to the longitudinal motion model, the quantized value information of the corresponding acceleration/brake control component under the conditions of the target acceleration information, current speed, and angle information;
  • the control instruction sending unit is configured to send a control instruction to the vehicle chassis control system according to the quantized value information of the accelerator/brake pedal control component.
  • a device for establishing a vehicle longitudinal motion model including:
  • Angle control unit for controlling the rotation of the vehicle steering wheel to the target sampling angle
  • the pedal opening control unit is used to gradually accelerate the vehicle to the highest safe driving speed under the target sampling angle condition by controlling the quantized value of the acceleration control component under the target sampling angle condition, and then by controlling the brake control component
  • the quantified value gradually decelerates the vehicle to a stop
  • the information collection unit is used to collect and record the corresponding relationship information between the quantized value of the acceleration/brake control component and the vehicle speed and acceleration according to a preset time interval during the acceleration/deceleration of the vehicle;
  • the model establishment unit is configured to establish the vehicle longitudinal motion model according to multiple sets of the corresponding relationships obtained under multiple target sampling angle conditions.
  • An automatic driving control device includes:
  • the model obtaining unit is used to obtain the longitudinal motion model of the vehicle
  • the longitudinal control target information receiving unit is used to receive the longitudinal control target information output by the motion control algorithm
  • a quantized value determining unit configured to determine quantized value information of the acceleration/brake control component according to the longitudinal motion model, the longitudinal control target information, and the current motion state information of the vehicle;
  • the instruction sending unit is configured to send a control instruction to the vehicle chassis control system according to the quantized value information of the acceleration/brake control component.
  • a computer system including:
  • One or more processors are One or more processors.
  • a memory associated with the one or more processors where the memory is used to store program instructions, and when the program instructions are read and executed by the one or more processors, perform the following operations:
  • the longitudinal motion model includes the corresponding relationship between the quantized value, speed, angle, and acceleration of the acceleration/brake control component of the vehicle;
  • the longitudinal motion model determine the quantized value information of the corresponding acceleration/brake control component under the conditions of the target acceleration information, current speed, and angle information;
  • a computer system including:
  • One or more processors are One or more processors.
  • a memory associated with the one or more processors where the memory is used to store program instructions, and when the program instructions are read and executed by the one or more processors, perform the following operations:
  • the vehicle Under the condition of the target sampling angle, the vehicle is gradually accelerated to the highest safe driving speed under the target sampling angle condition by controlling the quantized value of the acceleration control component, and then the vehicle is gradually decelerated to a stop by controlling the quantized value of the braking control component ;
  • the vehicle longitudinal motion model is established according to multiple sets of the corresponding relationships obtained under multiple target sampling angle conditions.
  • the longitudinal motion model can be created through specific data collection during the specific control of the vehicle driving process.
  • the control operation in the specific acquisition process may include: after controlling the vehicle to obtain the target sampling angle, gradually changing the accelerator/brake pedal opening to change the speed and acceleration of the vehicle, and in the process, recording the acceleration/ The corresponding relationship among brake pedal opening, speed, angle, and acceleration is used to describe the longitudinal motion model of the vehicle.
  • the longitudinal motion model of the vehicle can be simplified, and the dependence on the parameters of the various modules in the vehicle can be reduced.
  • the accuracy will be relatively high.
  • the model can be corrected in the actual automatic driving process, so that the final model of each vehicle can more accurately reflect the situation of the vehicle itself, reduce the impact of slight inconsistencies between different vehicles, and further improve accuracy Sex.
  • Figure 1 is a schematic diagram of a system provided by an embodiment of the present application.
  • FIG. 2 is a flowchart of the first method provided by an embodiment of the present application.
  • FIG. 3 is a flowchart of a second method provided by an embodiment of the present application.
  • Figure 4 is a flowchart of a third method provided by an embodiment of the present application.
  • Figure 5 is a schematic diagram of a first device provided by an embodiment of the present application.
  • Figure 6 is a schematic diagram of a second device provided by an embodiment of the present application.
  • Figure 7 is a schematic diagram of a third device provided by an embodiment of the present application.
  • Fig. 8 is a schematic diagram of a computer system provided by an embodiment of the present application.
  • the main purpose is to determine the quantized value information of the acceleration/brake control component based on the target acceleration calculated by the algorithm, combined with the current speed of the vehicle, steering wheel angle and other information (for example, in the vehicle In the case of an acceleration/brake pedal, the quantized value can specifically refer to the opening of the pedal, etc.), so that the quantized value information of the acceleration/brake control component is sent to the chassis control system of the vehicle, and the chassis control system Perform corresponding operations to achieve the purpose of accelerating or decelerating the vehicle by changing the acceleration.
  • the longitudinal motion model can be simplified as the quantized value information of the acceleration/brake control component, speed, angle (steering wheel angle or front wheel angle, etc.), acceleration information in the four dimensions.
  • the information in the other dimension can be determined when the information in the three dimensions is known.
  • the specific motion control algorithm will calculate the target acceleration, that is, the vehicle needs to be accelerated or decelerated according to this acceleration; in addition, according to the vehicle's positioning system, etc.
  • the current speed information of the vehicle can be obtained; in addition, the vehicle can also feed back its current angle information in real time, that is, the acceleration, speed, and angle are all known. Therefore, the corresponding relationship can be determined according to the corresponding relationship in the model.
  • the quantitative value information of the acceleration/brake pedal control component is not limited to, the acceleration/brake pedal control component.
  • a data acquisition system can be provided in a system such as a computing platform installed in a vehicle.
  • the computing platform may specifically include a perception module (obtaining perception of the vehicle itself and the surrounding environment according to various sensors), a path planning module, a motion control module, etc., which are mainly used to automatically plan the driving path of the vehicle.
  • the data acquisition system in the embodiment of the present application can run on the automatic control computing platform of the vehicle, and can generate a specific control message through the motion control module of the computing platform, and the control message can be sent through the CAN bus.
  • the vehicle chassis control system can control and respond to specific control messages and drive the vehicle to move in a prescribed manner.
  • the vehicle control system can also feed back some data to the data acquisition system, including speed, acceleration, etc.
  • the specific speed, acceleration and other information can also be learned through the vehicle's positioning and navigation system.
  • the data acquisition system can control the vehicle to execute the corresponding movement according to the pre-design requirements, and collect and record the quantitative value information, speed, angle, acceleration, etc. of specific acceleration/brake control components during the movement. . Later, the corresponding longitudinal motion model can be established according to the recorded information.
  • the longitudinal motion model of the vehicle can be simplified to express the correspondence between the quantized value information, speed, angle, and acceleration of the acceleration/brake control component. Therefore, it can be compared Simple; in addition, because the above model can be created by collecting data in the specific motion state of the vehicle, it also has higher accuracy. In addition, it can also be achieved in the specific automatic driving process. The information in is corrected to further improve accuracy.
  • the first embodiment first provides a method for establishing a longitudinal motion model of a vehicle from the perspective of a data acquisition system. See Fig. 2, including:
  • the solution adopted in the embodiment of this application may be: First, set several specific steering wheel angles as the target sampling angles, and then collect the correspondence between the quantized value information of the acceleration/brake control component and the vehicle speed and acceleration under the conditions of each sampling angle.
  • each target sampling angle can be set as the current angle condition, and then information such as speed, acceleration, and quantized value information of acceleration/brake control components can be collected. Therefore, after the collection is specifically started, the steering wheel of the vehicle can be controlled to rotate to the target sampling angle first, and the steering wheel angle will remain unchanged until the collection operation under this condition is completed. It should be noted that since the data acquisition system is set in the automatic driving computing platform, specific control messages can be sent directly through the motion control module of the computing platform, etc., to complete the control of the rotation angle of the vehicle steering wheel.
  • the control message sent by the specific data acquisition system to the vehicle chassis control system through the motion control module may include the steering wheel angle and the quantized value information of the acceleration/brake control component.
  • the quantized value information of the acceleration/brake control component can be controlled to change the driving speed and acceleration of the vehicle under the angle condition.
  • the initial opening degree of the accelerator pedal may be 0%, that is, the initial state of the vehicle may be a stopped state, or may be other states such as 10%. You can gradually increase the opening of the accelerator pedal through the preset step value, knowing that the speed of the vehicle has reached the maximum safe speed under the current steering wheel angle.
  • the maximum safe speed of the vehicle at a specific steering wheel angle may be preset.
  • the lateral acceleration of the vehicle should be less than 3m/s 2 , that is to say, in order to ensure the safe driving of the vehicle, the lateral acceleration of the vehicle should be controlled regardless of the specific steering wheel angle It is lower than the above threshold, otherwise dangers such as rollover may occur.
  • the lateral acceleration is caused by the arc motion of the vehicle during turning. This acceleration is related to the specific vehicle speed and the corresponding steering wheel angle. Therefore, when the steering wheel angle is known, it can be According to the maximum value of lateral acceleration, calculate the maximum speed that the vehicle can travel to under the corner condition, and use this speed as the maximum safe speed under the corresponding corner condition.
  • the relationship between the specific lateral acceleration, speed, and steering wheel angle can be calculated according to the specific transmission ratio, wheelbase, etc., which will not be detailed here.
  • the maximum safe speed allowed under specific steering wheel angle conditions can be correspondingly calculated.
  • the maximum safe speed can be set, for example, it can be 30m/s, or other values can be set according to the vehicle's own configuration and performance, etc.
  • the vehicle when the steering wheel angle is 100% of the full scale, the vehicle can only be accelerated to 3m/s by controlling the opening of the accelerator pedal. After reaching the maximum safe speed, it will Decelerate by controlling the opening of the brake pedal until the vehicle stops and the vehicle speed is zero. Similarly, when the steering wheel angle is 50% of the full scale, the vehicle can be accelerated to 5m/s by controlling the opening of the accelerator pedal. After reaching the maximum safe speed, it will pass the control The opening of the brake pedal decelerates until the vehicle stops, the vehicle speed is 0, and so on.
  • S203 During the acceleration/deceleration of the vehicle, collect and record the corresponding relationship information between the quantized value information of the acceleration/brake control component and the vehicle speed and acceleration according to a preset time interval;
  • the quantized value information of the acceleration/brake control component and the vehicle speed and acceleration can also be collected and recorded at preset time intervals. Correspondence information between.
  • the specific time interval can be determined according to actual accuracy requirements.
  • the corresponding time interval under different steering wheel angle conditions can also be different. For example, in specific implementation, it can be collected every 100ms, and so on.
  • the quantized value information about the acceleration/brake control component and the vehicle speed, acceleration and other information can be fed back by the vehicle system, or the speed, acceleration and other information can also be obtained through the vehicle's positioning system.
  • the specific correspondence can be recorded in the form of a table.
  • the recorded correspondence can be as shown in Table 1:
  • Correspondences under various other angle conditions can also be collected in the above-mentioned manner. Specifically, after the collection process under one of the angle conditions is completed, the steering wheel of the vehicle can be controlled to turn to the next target collection angle, and then repeat this step 203. Obtain a data collection result under a new angle condition. For example, the final collection results can be as shown in Table 2:
  • a vehicle longitudinal motion model can be established according to these corresponding relationships.
  • the specific vehicle longitudinal motion model can be expressed by the corresponding relationship in the four dimensions of the quantized value information of the acceleration/brake control component, speed, acceleration, and angle.
  • the collected data is equidistant in the time dimension, but due to the existence of acceleration, it will not be in other dimensions. Isometric. Therefore, in order to better use the longitudinal motion model for automatic driving control and remove the information in the dimension of sampling time in the collected data, it is also possible to target multiple sets of corresponding sets obtained under the condition of multiple target sampling angles.
  • the relationship is divided from the speed dimension according to the preset speed interval to form speed, angle (steering wheel angle, or front wheel angle), quantitative value information of acceleration/brake control components, and three-dimensional calibration table of acceleration (also That is, the information in another dimension can be calibrated by the information in any three dimensions), for example, in the velocity dimension, segmentation is performed at an interval of 0.1m/s, so that the final model can be as shown in Table 3:
  • the acceleration information in the corresponding relationship can also be filtered to remove the effects of burrs.
  • a specific longitudinal motion model is established, and then the longitudinal motion model can be used.
  • the motion model performs specific automatic driving control of the vehicle. It should be noted that in the embodiments of this application, data is collected during the driving process of a specific vehicle, and the corresponding longitudinal motion model is generated. In practical applications, since vehicles of the same brand, model, and style are The longitudinal motion parameters usually have strong consistency. Therefore, one of the vehicles can be used as a sample for data collection. Of course, in order to be more accurate, you can also select multiple vehicles of the same brand, model, and style to collect separately. After averaging, establish a longitudinal motion model, and so on. After the longitudinal motion model is established, it can be provided to the vehicle's specific autonomous driving calculation platform for use.
  • the automatic driving computing platform can usually be installed in the vehicle system. Through the perception of the vehicle and surrounding vehicles and the environment, it can plan the path of the automatic driving, and generate the corresponding control message, which is provided to the vehicle chassis through the CAN vertical.
  • the control system controls the movement of the vehicle.
  • the specific longitudinal motion model can be used to realize the generation of the control message for the longitudinal control.
  • the specific motion control module will generate two control messages, one is the acceleration, and the other is the angle.
  • the acceleration changes the vehicle’s Speed, change the direction of the vehicle through the angle. Among them, for longitudinal control, only the change of acceleration is concerned.
  • the acceleration information can be obtained by calculation of a specific motion control module, that is, through the perception of the vehicle itself, the surrounding environment, etc., it is determined that the vehicle needs to pass A certain acceleration changes the current speed, and the value of the acceleration can be calculated.
  • the current speed information of the vehicle can also be obtained through the positioning and navigation system of the vehicle.
  • the current angle information of the vehicle can also be obtained through the feedback information of the vehicle chassis control system. . In this way, the information equivalent to the three dimensions of acceleration, speed, and rotation angle is known. Therefore, it is only necessary to calibrate the quantitative value information of the specific acceleration/brake control component through the three-dimensional calibration table established above. .
  • the established correspondence relationship only includes the above-mentioned target sampling angle conditions, different speeds, different accelerations, and different accelerations.
  • the specific angle during the actual driving of the vehicle may be any value within the full-scale range. Therefore, during the specific calibration, the actual angle value can also be converted in the three-dimensional calibration table.
  • the steering wheel angle of the vehicle at a certain moment is 80% of the full range
  • the speed is 1.2m/s
  • the acceleration is -0.1m/s 2.
  • the steering wheel angle is full range because there is no specific three-dimensional calibration table.
  • the acceleration is the accelerator opening information / brake pedal -0.1m / s 2 corresponds, therefore, needs to be converted.
  • the acceleration/acceleration corresponding to a speed of 1.2m/s and an acceleration of -0.1m/s 2
  • the first quantized value information of the brake control component, and the second quantized value of the acceleration/brake control component corresponding to a speed of 1.2m/s and an acceleration of -0.1m/s 2 when the steering wheel angle is 50% of the full scale information.
  • the speed and acceleration are fixed, there is a linear relationship between the steering wheel angle and the quantized value information of the acceleration/brake control component.
  • the start and end points of the interval can be combined Acceleration/brake pedal opening information, calculated that when the steering wheel angle is 80% of the full scale, the third quantification of the acceleration/brake control component corresponding to a speed of 1.2m/s and an acceleration of -0.1m/s 2 Value information. Then, according to the third quantized value information, a corresponding control message can be sent to the vehicle chassis control system to control the vehicle to obtain a corresponding acceleration.
  • the acceleration information of the vehicle will change.
  • the actual angular velocity of the vehicle should be exactly the same as the acceleration calculated in the previous calculation platform.
  • data collection may only be carried out on some vehicles, and the specific modeling results are applied to multiple different vehicles of the same brand, style, and model.
  • the actual acceleration may be slightly different.
  • acceleration information actually obtained by a specific vehicle can also be obtained, and the longitudinal motion model in the current vehicle can be corrected.
  • the corrected value can be used for control to improve accuracy. In other words, in this way, if there are inconsistent parameters between different vehicles, they can be corrected in the actual automatic driving process. Through this correction, there will be some differences between the longitudinal motion models on each vehicle. The difference is more suitable for the actual situation of the specific vehicle itself, and more precise control results can be obtained.
  • the longitudinal motion model can be created through specific data collection during the specific control of the vehicle driving process.
  • the control operation in the specific acquisition process may include: after controlling the vehicle to obtain the target sampling angle, gradually changing the quantized value of the acceleration/brake control component, changing the speed and acceleration of the vehicle, and in the process, recording the vehicle's
  • the corresponding relationship between the quantized value, speed, angle, and acceleration of the acceleration/brake control component is used to describe the longitudinal motion model of the vehicle.
  • the longitudinal motion model of the vehicle can be simplified, and the dependence on the parameters of the various modules in the vehicle can be reduced.
  • the accuracy will be relatively high.
  • the vehicle chassis control system responds to the quantized value information of the acceleration/brake control component under the target speed and angle during the automatic driving control process using the above-mentioned longitudinal control model. And use the actual acceleration information to correct the vehicle longitudinal motion model.
  • the model can be corrected in the actual automatic driving process, so that the final model of each vehicle can more accurately reflect the situation of the vehicle itself, reduce the impact of slight inconsistencies between different vehicles, and further improve accuracy Sex.
  • the second embodiment corresponds to the first embodiment.
  • an automatic driving control method is provided. Referring to FIG. 3, the method may specifically include:
  • S301 Obtain a vehicle longitudinal motion model, where the longitudinal motion model includes the corresponding relationship between the quantized value, speed, angle, and acceleration of the acceleration/brake control component of the vehicle;
  • the longitudinal motion model is created by data collected in the process of controlling the driving of the vehicle, where the control of the vehicle during the collection process includes: obtaining the target sampling angle in the control vehicle Later, by gradually changing the quantized value of the acceleration/brake control component, the speed and acceleration of the vehicle are changed.
  • S302 Receive target acceleration information output by the motion control algorithm, and obtain current speed and angle information of the vehicle;
  • the specific target acceleration information can be calculated by the path planning module, motion control module, etc. in the computing platform according to the current situation of the vehicle itself and the surrounding environment, that is, the vehicle needs to be controlled to change to the target acceleration to avoid Collide with other vehicles, or get better driving space and performance for the current vehicle, etc.
  • S303 According to the longitudinal motion model, determine the corresponding quantized value of the acceleration/brake pedal control component under the conditions of the target acceleration information, current speed, and angle information;
  • the longitudinal motion model includes the corresponding relationship between acceleration/brake pedal opening, speed, and acceleration under multiple preset target sampling angle conditions.
  • the target sampling angles can also be used as nodes to divide into multiple angle intervals;
  • the acceleration/brake corresponding to the target acceleration information and the current speed can be determined at the current angle
  • the third opening information of the pedal is the first opening degree information, the second opening degree information and the preset linear function relationship information.
  • the linear function relationship information is determined according to the following characteristics: when the speed and acceleration are fixed, there is a linear relationship between the angle of the vehicle and the opening information of the acceleration/brake pedal.
  • S304 Send a control instruction to the vehicle chassis control system according to the quantized value information of the acceleration/brake control component.
  • the vehicle chassis control system After sending a control instruction to the vehicle chassis control system, the vehicle chassis control system can respond to the instruction and change the acceleration of the vehicle by changing the accelerator/brake pedal opening information.
  • the actual acceleration information obtained by the vehicle after the vehicle chassis control system responds to the acceleration/brake pedal opening information can also be obtained; the actual acceleration information is used to correct the vehicle longitudinal motion model.
  • the longitudinal motion model of the vehicle is expressed by the corresponding relationship between the quantized value information, speed, angle, and acceleration of the acceleration/brake control component of the vehicle.
  • the specific longitudinal motion model is There may also be multiple expression modes. Therefore, an embodiment of the present application also provides an automatic driving control method. Referring to FIG. 4, the method may specifically include:
  • S402 Receive longitudinal control target information output by the motion control algorithm
  • the longitudinal control target usually may also be specific target acceleration information, or may also include speed information.
  • S403 Determine the quantized value information of the acceleration/brake control component according to the longitudinal motion model, the longitudinal control target information, and the current motion state information of the vehicle;
  • the current motion state information can include the current speed, vehicle angle, and so on.
  • S404 Send a control instruction to the vehicle chassis control system according to the quantized value information of the acceleration/brake control component.
  • this embodiment of the present application also provides a device for establishing a longitudinal motion model of a vehicle.
  • the device may specifically include:
  • the angle control unit 501 is used to control the rotation of the vehicle steering wheel to the target sampling angle
  • the speed control unit 502 is used to control the quantized value of the acceleration control component to gradually accelerate the vehicle to the highest safe driving speed under the target sampling angle under the condition of the target sampling angle, and then control the quantization of the brake control component Value slows the vehicle to a stop gradually;
  • the information collection unit 503 is configured to collect and record the corresponding relationship information between the quantized value information of the acceleration/brake control component and the vehicle speed and acceleration according to a preset time interval during the acceleration/deceleration of the vehicle;
  • the model establishment unit 504 is configured to establish the vehicle longitudinal motion model according to multiple sets of the corresponding relationships obtained under multiple target sampling angle conditions.
  • the device may also include:
  • the segmentation unit is configured to segment the multiple sets of the corresponding relationships obtained under the conditions of the multiple target sampling angles, respectively, from the speed dimension, according to a preset speed interval.
  • the filtering unit is configured to perform filtering processing on the acceleration information in the corresponding relationship.
  • the actual acceleration obtaining unit is used to obtain the quantitative value information of the acceleration/brake control component of the vehicle chassis control system in response to the target speed and angle during the automatic driving control of the vehicle using the longitudinal motion model, Actual acceleration information obtained by the vehicle;
  • the correction unit is configured to use the actual acceleration information to correct the longitudinal motion model of the vehicle.
  • the embodiment of the present application also provides an automatic driving control device.
  • the device may specifically include:
  • the model obtaining unit 601 is configured to obtain a longitudinal motion model of a vehicle, wherein the longitudinal motion model includes the corresponding relationship between the quantized value, speed, angle, and acceleration of the acceleration/brake control component of the vehicle;
  • the information obtaining unit 602 is configured to receive target acceleration information output by the motion control algorithm, and obtain the current speed and angle information of the vehicle;
  • the quantized value information determining unit 603 is configured to determine, according to the longitudinal motion model, the quantized value information of the corresponding acceleration/brake control component under the conditions of the target acceleration information, current speed and angle information;
  • the control instruction sending unit 604 is configured to send a control instruction to the vehicle chassis control system according to the quantized value information of the acceleration/brake control component.
  • the longitudinal motion model may be created by data collected during the process of controlling the driving of the vehicle, where the control of the vehicle during the collection process includes: after controlling the vehicle to obtain the target sampling angle, gradually changing the acceleration /The quantized value information of the brake control component changes the speed and acceleration of the vehicle.
  • the longitudinal motion model includes the corresponding relationship between the quantized value information, speed, and acceleration of the acceleration/brake control component under multiple preset target sampling angle conditions.
  • the opening information determining unit may specifically include:
  • An interval division subunit for dividing the target sampling angle into multiple angle intervals
  • the first determining subunit is used to determine the target acceleration information, the first quantized value information of the acceleration/brake control component corresponding to the current speed, and the second target sampling angle under the condition of the first target sampling angle Under conditions, the target acceleration information and the second quantized value information of the acceleration/brake control component corresponding to the current speed;
  • the second determining subunit is configured to determine the acceleration/acceleration corresponding to the target acceleration information and the current speed at the current angle according to the first quantization value information, the second quantization value information, and preset linear function relationship information.
  • the third quantized value information of the brake control component is configured to determine the acceleration/acceleration corresponding to the target acceleration information and the current speed at the current angle according to the first quantization value information, the second quantization value information, and preset linear function relationship information.
  • the linear function relationship information is determined according to the following characteristics: when the speed and acceleration are fixed, there is a linear relationship between the angle of the vehicle and the quantized value information of the acceleration/brake control component.
  • the device may also include:
  • An actual acceleration information obtaining unit configured to obtain actual acceleration information obtained by the vehicle after the vehicle chassis control system responds to the quantized value information of the acceleration/brake control component;
  • the correction unit is configured to use the actual acceleration information to correct the longitudinal motion model of the vehicle.
  • the embodiment of the present application also provides an automatic driving control device.
  • the device may include:
  • the model obtaining unit 701 is configured to obtain a longitudinal motion model of the vehicle
  • the longitudinal control target information receiving unit 702 is configured to receive the longitudinal control target information output by the motion control algorithm
  • the quantized value determining unit 703 is configured to determine the quantized value information of the acceleration/brake control component according to the longitudinal motion model, the longitudinal control target information, and the current motion state information of the vehicle;
  • the control instruction sending unit 704 is configured to send a control instruction to the vehicle chassis control system according to the quantized value information of the acceleration/brake control component.
  • an embodiment of the present application also provides a computer system, including:
  • One or more processors are One or more processors.
  • a memory associated with the one or more processors where the memory is used to store program instructions, and when the program instructions are read and executed by the one or more processors, perform the following operations:
  • the longitudinal motion model includes the corresponding relationship between the quantized value, speed, angle, and acceleration of the acceleration/brake control component of the vehicle;
  • the longitudinal motion model determine the quantized value information of the corresponding acceleration/brake control component under the conditions of the target acceleration information, current speed, and angle information;
  • One or more processors are One or more processors.
  • a memory associated with the one or more processors where the memory is used to store program instructions, and when the program instructions are read and executed by the one or more processors, perform the following operations:
  • the vehicle Under the condition of the target sampling angle, the vehicle is gradually accelerated to the highest safe driving speed under the target sampling angle condition by controlling the quantized value of the acceleration control component, and then the vehicle is gradually decelerated to a stop by controlling the quantized value of the braking control component ;
  • the vehicle longitudinal motion model is established according to multiple sets of the corresponding relationships obtained under multiple target sampling angle conditions.
  • FIG. 8 exemplarily shows the architecture of the computer system, which may specifically include a processor 810, a video display adapter 811, a disk drive 812, an input/output interface 813, a network interface 814, and a memory 820.
  • the processor 810, the video display adapter 811, the disk drive 812, the input/output interface 813, the network interface 814, and the memory 820 may be communicatively connected through the communication bus 830.
  • the processor 810 may be implemented by a general CPU (Central Processing Unit, central processing unit), a microprocessor, an application specific integrated circuit (Application Specific Integrated Circuit, ASIC), or one or more integrated circuits. Perform relevant procedures to realize the technical solutions provided in this application.
  • a general CPU Central Processing Unit, central processing unit
  • a microprocessor central processing unit
  • ASIC Application Specific Integrated Circuit
  • the memory 820 may be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory, random access memory), static storage device, dynamic storage device, etc.
  • the memory 820 may store an operating system 821 used to control the operation of the computer system 800, and a basic input output system (BIOS) 822 used to control low-level operations of the computer system 800.
  • BIOS basic input output system
  • a web browser 823, a data storage management system 824, and an automatic driving processing system 825 can also be stored.
  • the foregoing automatic driving processing system 825 may be an application program that specifically implements the foregoing steps in the embodiment of the present application. In short, when the technical solution provided by the present application is implemented through software or firmware, the related program code is stored in the memory 820 and is called and executed by the processor 810.
  • the input/output interface 813 is used to connect an input/output module to realize information input and output.
  • the input/output/module can be configured in the device as a component (not shown in the figure), or can be connected to the device to provide corresponding functions.
  • the input device may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and an output device may include a display, a speaker, a vibrator, an indicator light, and the like.
  • the network interface 814 is used to connect to a communication module (not shown in the figure) to implement communication interaction between the device and other devices.
  • the communication module can realize communication through wired means (such as USB, network cable, etc.), or through wireless means (such as mobile network, WIFI, Bluetooth, etc.).
  • the bus 830 includes a path for transmitting information between various components of the device (for example, the processor 810, the video display adapter 811, the disk drive 812, the input/output interface 813, the network interface 814, and the memory 820).
  • various components of the device for example, the processor 810, the video display adapter 811, the disk drive 812, the input/output interface 813, the network interface 814, and the memory 820.
  • the computer system 800 can also obtain information about specific receiving conditions from the virtual resource object receiving condition information database 841 for condition judgment, and so on.
  • the above device only shows the processor 810, the video display adapter 811, the disk drive 812, the input/output interface 813, the network interface 814, the memory 820, the bus 830, etc., in the specific implementation process, the The device may also include other components necessary for normal operation.
  • the above-mentioned device may also include only the components necessary for implementing the solution of the present application, and not necessarily all the components shown in the figure.

Landscapes

  • Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Human Computer Interaction (AREA)
  • Regulating Braking Force (AREA)
  • Control Of Driving Devices And Active Controlling Of Vehicle (AREA)

Abstract

一种建立车辆纵向运动模型的方法、装置及计算机系统,该方法包括:控制车辆方向盘旋转至目标采样角度;在该目标采样角度的条件下,通过控制加速控制部件的量化值信息将车辆逐渐加速至该目标采样角度条件下的最高安全行车速度,再通过控制制动控制部件的量化值信息将车辆逐渐减速至停止;在所述车辆加速/减速的过程中,按照预置的时间间隔采集并记录加速/制动控制部件的量化值信息与车辆速度、加速度之间的对应关系信息;根据多个目标采样角度条件下获得的多组所述对应关系,建立所述车辆纵向运动模型。通过简单、高精度的纵向运动模型,实现更精确的自动驾驶控制。

Description

建立车辆纵向运动模型的方法、装置及计算机系统
本申请要求2019年03月05日递交的申请号为201910162589.8、发明名称为“建立车辆纵向运动模型的方法、装置及计算机系统”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
本申请涉及自动驾驶技术领域,特别是涉及建立车辆纵向运动模型的方法、装置及计算机系统。
背景技术
在车辆自动驾驶控制系统中,对车辆的控制主要包括对行车速度以及方向上的控制。其中,关于车速以及本车与前后车或障碍物距离的自动控制属于一种纵向控制。横向控制指垂直于运动方向上的控制,对于汽车而言也就是转向控制。
其中,纵向控制问题可归结为对发动机输出和刹车的控制。在传统的实现方案中,通过各种发动机模型、汽车运行模型和刹车过程模型与不同的控制器算法结合,建立发动机和汽车运动过程的近似线形模型,在此基础上设计控制器。这种方法实现的控制,需要准确的获知车辆的驱动器、制动器等多个主要模块的参数,因此,参数依赖性大及模型误差较大,所以精度差、适应性差。
因此,如何实现更精确的自动驾驶控制,成为需要本领域技术人员解决的技术问题。
发明内容
本申请提供了建立车辆纵向运动模型的方法、装置及计算机系统,能够通过更简单、更高精度的纵向运动模型,实现更精确的自动驾驶控制。
本申请提供了如下方案:
一种自动驾驶控制方法,包括:
获得车辆纵向运动模型,其中,所述纵向运动模型中包括车辆的加速/制动控制部件的量化值信息、速度、角度、加速度之间的对应关系;
接收运动控制算法输出的目标加速度信息,并获得车辆的当前速度以及角度信息;
根据所述纵向运动模型,确定在所述目标加速度信息、当前速度以及角度信息的条件下,对应的加速/制动控制部件的量化值信息;
根据所述加速/制动控制部件的量化值信息信息向车辆底盘控制系统发送控制指令。
一种建立车辆纵向运动模型的方法,包括:
控制车辆方向盘旋转至目标采样角度;
在该目标采样角度的条件下,通过控制加速控制部件的量化值将车辆逐渐加速至该目标采样角度条件下的最高安全行车速度,再通过控制制动控制部件的量化值将车辆逐渐减速至停止;
在所述车辆加速/减速的过程中,按照预置的时间间隔采集并记录加速/制动控制部件的量化值与车辆速度、加速度之间的对应关系信息;
根据多个目标采样角度条件下获得的多组所述对应关系,建立所述车辆纵向运动模型。
一种自动驾驶控制方法,包括:
获得车辆纵向运动模型;
接收运动控制算法输出的纵向控制目标信息;
根据所述纵向运动模型、所述纵向控制目标信息以及车辆的当前运动状态信息,确定加速/制动控制部件的量化值信息;
根据所述加速/制动控制部件的量化值信息信息向车辆底盘控制系统发送控制指令。
一种自动驾驶控制装置,包括:
模型获得单元,用于获得车辆纵向运动模型,其中,所述纵向运动模型中包括车辆的加速/制动控制部件的量化值、速度、角度、加速度之间的对应关系;
信息获得单元,用于接收运动控制算法输出的目标加速度信息,并获得车辆的当前速度以及角度信息;
量化值信息确定单元,用于根据所述纵向运动模型,确定在所述目标加速度信息、当前速度以及角度信息的条件下,对应的加速/制动控制部件的量化值信息;
控制指令发送单元,用于根据所述加速/制动踏板控制部件的量化值信息向车辆底盘控制系统发送控制指令。
一种建立车辆纵向运动模型的装置,包括:
角度控制单元,用于控制车辆方向盘旋转至目标采样角度;
踏板开度控制单元,用于在该目标采样角度的条件下,通过控制加速控制部件的量化值将车辆逐渐加速至该目标采样角度条件下的最高安全行车速度,再通过控制制动控制部件的量化值将车辆逐渐减速至停止;
信息采集单元,用于在所述车辆加速/减速的过程中,按照预置的时间间隔采集并记录加速/制动控制部件的量化值与车辆速度、加速度之间的对应关系信息;
模型建立单元,用于根据多个目标采样角度条件下获得的多组所述对应关系,建立所述车辆纵向运动模型。
一种自动驾驶控制装置,包括:
模型获得单元,用于获得车辆纵向运动模型;
纵向控制目标信息接收单元,用于接收运动控制算法输出的纵向控制目标信息;
量化值确定单元,用于根据所述纵向运动模型、所述纵向控制目标信息以及车辆的当前运动状态信息,确定加速/制动控制部件的量化值信息;
指令发送单元,用于根据所述加速/制动控制部件的量化值信息信息向车辆底盘控制系统发送控制指令。
一种计算机系统,包括:
一个或多个处理器;以及
与所述一个或多个处理器关联的存储器,所述存储器用于存储程序指令,所述程序指令在被所述一个或多个处理器读取执行时,执行如下操作:
获得车辆纵向运动模型,其中,所述纵向运动模型中包括车辆的加速/制动控制部件的量化值、速度、角度、加速度之间的对应关系;
接收运动控制算法输出的目标加速度信息,并获得车辆的当前速度以及角度信息;
根据所述纵向运动模型,确定在所述目标加速度信息、当前速度以及角度信息的条件下,对应的加速/制动控制部件的量化值信息;
根据所述加速/制动控制部件的量化值信息向车辆底盘控制系统发送控制指令。
一种计算机系统,包括:
一个或多个处理器;以及
与所述一个或多个处理器关联的存储器,所述存储器用于存储程序指令,所述程序指令在被所述一个或多个处理器读取执行时,执行如下操作:
控制车辆方向盘旋转至目标采样角度;
在该目标采样角度的条件下,通过控制加速控制部件的量化值将车辆逐渐加速至该目标采样角度条件下的最高安全行车速度,再通过控制制动控制部件的量化值将车辆逐渐减速至停止;
在所述车辆加速/减速的过程中,按照预置的时间间隔采集并记录加速/制动控制部件 的量化值与车辆速度、加速度之间的对应关系信息;
根据多个目标采样角度条件下获得的多组所述对应关系,建立所述车辆纵向运动模型。
根据本申请提供的具体实施例,本申请公开了以下技术效果:
通过本申请实施例,可以在具体控制车辆行驶过程中,通过具体的数据采集创建纵向运动模型。其中,具体采集过程中的控制操作可以包括:在控制车辆获得目标采样角度后,通过逐渐改变加速/制动踏板开度,改变车辆的速度以及加速度,并在此过程中,记录车辆的加速/制动踏板开度、速度、角度、加速度之间的对应关系,据此对车辆的纵向运动模型进行描述。这样,可以简化车辆的纵向运动模型,降低对车辆内部各模块参数的依赖,并且,由于是在车辆具体行驶过程中进行的数据采集,因此,精度也会比较高。
另外,还可以在具体利用上述纵向控制模型进行自动驾驶控制的过程中,获得车辆底盘控制系统在目标速度、角度情况下,响应加速/制动踏板开度信息后,车辆获得的实际加速度信息,并利用所述实际加速度信息对所述车辆纵向运动模型进行校正。也就是说,可以实现实际自动驾驶过程中对模型的校正,使得最终每个车辆的模型能够更准确的反应车辆自身的情况,降低不同车辆之间的轻微不一致性带来的影响,进一步提高精确性。
当然,实施本申请的任一产品并不一定需要同时达到以上所述的所有优点。
附图说明
为了更清楚地说明本申请实施例或现有技术中的技术方案,下面将对实施例中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本申请的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他的附图。
图1是本申请实施例提供的系统的示意图;
图2是本申请实施例提供的第一方法的流程图;
图3是本申请实施例提供的第二方法的流程图;
图4是本申请实施例提供的第三方法的流程图;
图5是本申请实施例提供的第一装置的示意图;
图6是本申请实施例提供的第二装置的示意图;
图7是本申请实施例提供的第三装置的示意图;
图8是本申请实施例提供的计算机系统的示意图。
具体实施方式
下面将结合本申请实施例中的附图,对本申请实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅仅是本申请一部分实施例,而不是全部的实施例。基于本申请中的实施例,本领域普通技术人员所获得的所有其他实施例,都属于本申请保护的范围。
在本申请实施例中,提供了更简单,但精确性更高的纵向运动模型,在此基础上实现对车辆的自动控制。其中,由于在纵向控制的过程中,主要的目的就是根据算法计算出的目标加速度,结合车辆当前的速度、方向盘角度等信息,确定出加速/制动控制部件的量化值信息(例如,在车辆具有加速/制动踏板的情况下,该量化值具体可以是指踏板的开度,等等),以便将加速/制动控制部件的量化值信息发送给车辆的底盘控制系统,由底盘控制系统执行对应的操作,以达到通过改变加速度使车辆加速或者减速的目的。因此,在本申请实施例中,就可以将纵向运动模型简化为加速/制动控制部件的量化值信息、速度、角度(方向盘转角或者前轮转角等)、加速度这四个维度上的信息之间的关系,只要能够预先获知上述几个方面的信息之间的对应关系,则可以在其中三个维度上的信息已知的情况下,确定出另一个维度上的信息。例如,具体在对车辆进行自动驾驶的运动控制的过程中,具体的运动控制算法会计算出目标加速度,也即,需要让车辆按照这个加速度进行加速或者减速;另外,根据车辆的定位系统等,还可以获知车辆的当前速度信息;再者车辆还可以实时反馈其当前的角度信息,也就是说,加速度、速度、角度都是已知的,因此,可以根据模型中的对应关系,确定出对应的加速/制动踏控制部件的量化值信息。
其中,为了建立上述纵向运动模型,可以在具体的车辆中进行数据采集,然后,根据采集到的数据进行模型的建立。具体实现时,如图1所示,可以在在车辆安装的计算平台等系统中,提供数据采集系统。其中,该计算平台中具体可以包括感知模块(根据各种传感器获得对车辆自身以及周围环境的感知)、路径规划模块、运动控制模块等,主要用于为车辆进行行驶路径的自动规划。而本申请实施例中的数据采集系统就可以运行在车辆的自动控制计算平台中,能够通过计算平台的运动控制模块生成具体的控制报文,该控制报文则可以通过CAN总线等方式,发送到车辆底盘控制系统,车辆底盘控制 系统则可以控制响应具体的控制报文,驱动车辆按照规定的方式进行运动。另外,车辆控制系统还可以向数据采集系统反馈一些数据,包括速度、加速度等,当然,具体的速度、加速度等信息,也可以通过车辆的定位导航系统等进行获知。这样,数据采集系统便可以根据预先的设计要求,控制车辆执行对应的运动,并在运动中对具体的加速/制动控制部件的量化值信息、速度、角度、加速度等进行采集,并进行记录。后续便可以根据记录的信息建立对应的纵向运动模型。
通过这种方式,将车辆的纵向运动模型可以简化为通过加速/制动控制部件的量化值信息、速度、角度、加速度这四个维度上的信息之间的对应关系进行表达,因此,可以比较简单;另外,由于上述模型可以在车辆具体的运动状态下通过对数据进行采集的方式而进行创建,因此,也具有更高的精确性,另外,还可以实现在具体自动驾驶过程中,对模型中的信息进行校正,进一步提高精确度。
下面对具体的实现方案进行详细介绍。
实施例一
该实施例一首先从数据采集系统的角度,提供了一种建立车辆纵向运动模型的方法,参见图2,包括:
S201:控制车辆方向盘旋转至目标采样角度;
具体实现时,由于在速度、加速度固定的情况下,方向盘转角与加速/制动踏板开度之间基本呈线性关系,因此,为了方便进行数据采集,本申请实施例中采用的方案可以是:首先设定几个特定的方向盘角度作为目标采样角度,然后,分别在各个采样角度条件下,分别对加速/制动控制部件的量化值信息与车辆速度、加速度之间的对应关系进行采集。例如,所述目标采样角度可以包括方向盘转角分为满量程的100%,50%,15%,2%,0。具体如,假设某车辆方向盘转角满量程为470度,传动比16,也即方向盘打满量程后,前轮的旋转角度是470/16=29.375度。则具体在采样时,可以在方向盘转角为470度、235度、70.5度、9.4度、0度的状态下,分别进行采样。
在设备了具体的目标采样角度后,便可以分别将各个目标采样角度设为当前的角度条件,然后进行速度、加速度、加速/制动控制部件的量化值信息等信息的采集。因此,在具体开始进行采集后,首先就可以控制车辆方向盘旋转至目标采样角度,在该条件下的采集操作结束之前,方向盘转角一直保持该角度不变。需要说明的是,由于数据采集系统设置在自动驾驶计算平台中,因此,可以直接通过计算平台的运动控制模块等,发送具体的控制报文,以完成对车辆方向盘旋转角度的控制。
S202:在该目标采样角度的条件下,通过控制加速控制部件的量化值将车辆逐渐加速至该目标采样角度条件下的最高安全行车速度,再通过控制制动控制部件的量化值将车辆逐渐减速至停止;
在将车辆的方向盘转角设置为某个角度后,可以控制通过控制加速控制部件的量化值将车辆逐渐加速至该目标采样角度条件下的最高安全行车速度,然后,再通过控制制动控制部件的量化值将车辆逐渐减速至停止。也就是说,具体的数据采集系统通过运动控制模块向车辆底盘控制系统发送的控制消息可以包括方向盘转角以及加速/制动控制部件的量化值信息这两个方面的信息。在初始状态下控制车辆方向盘旋转到某个角度后,便可以通过控制加速/制动控制部件的量化值信息,来改变车辆在该角度条件下的行驶速度以及加速度。
例如,具体实现时,可以是加速踏板的初始开度可以是0%,也即,车辆初始状态可以是停止状态,或者,也可以是10%等其他的状态。可以通过预置的步进值,逐渐增加加速踏板的开度,知道车辆的速度达到当前方向盘转角下的最大安全速度。
具体的,关于车辆在具体方向盘转角下的最大安全速度可以是预先设定的。根据人类驾驶经验和车辆稳定性,车辆行驶时的侧向加速度应小于3m/s 2,也就是说,为了保证车辆的安全行驶,无论具体的方向盘转角为多少,都应该控制车辆的侧向加速度低于上述阈值,否则可能会发生侧翻等危险。而侧向加速度是由于车辆正在转弯过程中做出的圆弧运动导致的,该加速度与具体的车辆行驶速度以及对应的方向盘转角是相关的,因此,在已知方向盘转角的情况下,就可以根据侧向加速度的最大值,计算出在该转角条件下车辆能够行驶到的最高速度,将该速度作为对应转角条件下的最大安全速度即可。
其中,关于具体的侧向加速度、速度、方向盘转角之间的关系,可以根据具体的传动比、轴距等进行计算,这里不进行详述。总之,在安全范围内允许的侧向加速度的最大值固定不变的情况下,由于预先设定了方向盘转角,因此,可以对应计算出具体方向盘转角条件下允许的最大安全速度。例如,方向盘转角分别为满量程的100%,50%,15%,2%时,对应的车辆最大安全速度分别为3m/s,5m/s,10m/s,20m/s。另外,在方向盘转角为0%时,可以对最大安全速度进行设定,例如,可以为30m/s,或者根据车辆自身的配置、性能等设定其他的数值,等等。
也就是说,在方向盘转角为满量程的100%的情况下,通过控制加速踏板的开度进行角度的过程中,只能将车辆加速至3m/s,在达到该最大安全速度后,将会通过控制制动踏板的开度进行减速,直到车辆停止,车辆速度为0。类似的,在方向盘转角为满量程 的50%的情况下,通过控制加速踏板的开度进行角度的过程中,可以将车辆加速至5m/s,在达到该最大安全速度后,将会通过控制制动踏板的开度进行减速,直到车辆停止,车辆速度为0,以此类推。
S203:在所述车辆加速/减速的过程中,按照预置的时间间隔采集并记录加速/制动控制部件的量化值信息与车辆速度、加速度之间的对应关系信息;
在通过控制加速/制动控制部件的量化值信息对车辆进行加速或者减速控制的过程中,还可以按照预置的时间间隔采集并记录加速/制动控制部件的量化值信息与车辆速度、加速度之间的对应关系信息。具体的时间间隔可以根据实际的精度需要而定,另外,不同的方向盘转角条件下对应的时间间隔也可以是不同的。例如,具体实现时,可以是每100ms采集一次,等等。
其中,关于加速/制动控制部件的量化值信息与车辆速度、加速度等信息,可以是由车辆系统反馈,或者,关于速度、加速度等信息,还可以通过车辆的定位系统进行获得。
具体可以通过表格的形式记录具体的对应关系,例如,在方向盘角度为满量程的100%的情况下,记录的对应关系可以如表1所示:
表1
Figure PCTCN2020076530-appb-000001
其他各种角度条件下的对应关系也可以分别通过上述方式进行采集,具体的,在其中一个角度条件下的采集过程结束后,可以控制车辆方向盘转到下一个目标采集角度,然后重复执行该步骤203,获得新的角度条件下的数据采集结果。例如,最终获得的采集结果可以如表2所示:
表2
Figure PCTCN2020076530-appb-000002
Figure PCTCN2020076530-appb-000003
S204:根据多个目标采样角度条件下获得的多组所述对应关系,建立所述车辆纵向运动模型。
在得到上述多个目标采样角度条件下获得的多组所述对应关系后,就可以根据这些对应关系建立车辆纵向运动模型。在本申请实施例中,具体的车辆纵向运动模型,就可以由加速/制动控制部件的量化值信息、速度、加速度、角度这四个维度上的对应关系来进行表达。
具体实现时,由于数据采集的过程中通常是按照一定的时间间隔进行采集,因此,采集到的数据在时间维度上是等距的,但是,由于加速度的存在,在其他维度上则会是不等距的。因此,为了能够更好的使用该纵向运动模型进行自动驾驶控制,并去掉采集数据中采样时间这一维度上的信息,还可以针对所述多个目标采样角度条件下获得的多组所述对应关系,分别从速度维度上,按照预置的速度间隔进行切分,形成速度、角度(方向盘转角,或前轮转角)、加速/制动控制部件的量化值信息、加速度的三维标定表(也即,可以通过其中任意三个维度上的信息对另一个维度上的信息进行标定),例如,在速度维度上进行间隔0.1m/s的切分,使得最终的模型可以如表3所示:
表3
Figure PCTCN2020076530-appb-000004
Figure PCTCN2020076530-appb-000005
另外,在具体实现时,还可以对所述对应关系中的加速度信息进行滤波处理,去除毛刺的影响在得到上述三维标定表后,就建立起了具体的纵向运动模型,进而就可以利用该纵向运动模型进行具体的车辆自动驾驶控制。需要说明的是,在本申请实施例中,是在具体车辆的行驶过程中进行数据采集,并生成对应的纵向运动模型,而在实际应用中,由于同一品牌、型号、款式的车辆,在具体的纵向运动参数方面通常具有比较强的一致性,因此,可以针对其中一个车辆作为样本进行数据采集,当然,为了更加准确,也可以选择同一品牌、型号、款式的多个车辆分别进行采集,取均值后建立纵向运动模型,等等。在建立起纵向运动模型后,就可以提供给车辆的具体自动驾驶计算平台来使用。
具体的,自动驾驶计算平台通常可以安装在车辆系统内,通过对车辆以及周边车辆、环境的感知,可以进行自动驾驶的路径规划,并生成对应的控制报文,通过CAN纵向等提供给车辆底盘控制系统,对车辆进行运动控制。在本申请实施例中,具体的纵向运动模型就可以用来实现纵向控制的控制报文的生成。具体的,计算平台中的路径规划模块规划出具体的行驶路径信息后,具体的运动控制模块会生成两个方面的控制报文,一个是加速度方面,另一个是角度方面,通过加速度改变车辆的速度,通过角度改变车辆的方向。其中,对于纵向控制而言,只关心加速度的变化,该加速度信息可以是由具体的运动控制模块的通过计算获得的,也即,通过对车辆本身、周边环境等的感知,确定出车辆需要通过一定的加速度改变当前速度,并且可以计算出该加速度的值。在获得目标加速度信息后,还可以通过车辆的定位导航系统等获得车辆的当前速度信息,另外,还可以通过车辆底盘控制系统等的反馈信息获得车辆当前的角度信息(方向盘转角或者前轮转角)。这样,相当于加速度、速度、转角这三个维度上的信息都是已知的,因此,只需要通过前文建立的三维标定表,标定出具体的加速/制动控制部件的量化值信息即可。
需要说明的是,在本申请实施例中,只是对几个特定的目标采样角度进行了数据采集,建立的对应关系中也只包括上述目标采样角度条件下,不同速度、不同加速度与不同加速/制动控制部件的量化值信息之间的对应关系。但是,在车辆实际行驶过程中具体的角度可能是满量程范围内的任意值,因此,具体在进行标定时,还可以根据实际的角度值,在所述三维标定表中进行换算。
例如,假设某时刻车辆的方向盘转角是满量程的80%,速度是1.2m/s,加速度是-0.1m/s 2,此时,由于具体的三维标定表中不存在方向盘转角是满量程的80%时,与速度1.2m/s,加速度是-0.1m/s 2对应的加速/制动踏板的开度信息,因此,需要进行换算。具体的,由于80%位于100%和50%之间,因此,可以首先查询出方向盘转角是满量程的100%时,与速度1.2m/s,加速度是-0.1m/s 2对应的加速/制动控制部件的第一量化值信息,以及方向盘转角是满量程的50%时,与速度1.2m/s,加速度是-0.1m/s 2对应的加速/制动控制部件的第二量化值信息。由于在速度、加速度固定的条件下,方向盘转角与加速/制动控制部件的量化值信息之间成线性变化关系,因此,只要再预先获知该线性变化函数,就可以结合上述区间起点以及终点处的加速/制动踏板的开度信息,计算出方向盘转角是满量程的80%时,与速度1.2m/s,加速度是-0.1m/s 2对应的加速/制动控制部件的第三量化值信息。然后,可以根据该第三量化值信息,向车辆底盘控制系统发送对应的控制报文,以控制车辆获得对应的加速度。
在车辆底盘系统响应具体的控制消息后,车辆的加速度信息会发生变化,在完全精确控制的情况下,车辆实际获得的角速度与之前计算平台中计算出的加速度应该是完全一致的。当然,在实际应用中,如前文所述,由于数据采集可能只是在部分车辆上进行,并将具体的建模结果应用到同品牌、款式、型号的多个不同车辆上,而不同的车辆之间虽然存在比较强的一致性,但是,也可能存在部分车辆的实际参数存在一些细小差异的情况,使得不同车辆在相同速度、角度的状态下,响应相同的加速/制动控制部件的量化值信息后,实际获得的加速度可能会略有差异。因此,在本申请实施例中,还可以获得具体车辆实际获得的加速度信息,并且可以对当前车辆中的纵向运动模型进行校正。后续再进行自动驾驶控制的过程中,便可以利用校正后的值进行控制,以提高精确性。也就是说,通过这种方式,不同车辆之间如果存在参数不一致的情况,则可以在实际的自动驾驶过程中进行校正,通过这种校正使得每个车辆上的纵向运动模型之间会存在一些差异,更适合具体车辆自身的实际情况,获得更精确的控制结果。
总之,通过本申请实施例,可以在具体控制车辆行驶过程中,通过具体的数据采集 创建纵向运动模型。其中,具体采集过程中的控制操作可以包括:在控制车辆获得目标采样角度后,通过逐渐改变加速/制动控制部件的量化值,改变车辆的速度以及加速度,并在此过程中,记录车辆的加速/制动控制部件的量化值、速度、角度、加速度之间的对应关系,据此对车辆的纵向运动模型进行描述。这样,可以简化车辆的纵向运动模型,降低对车辆内部各模块参数的依赖,并且,由于是在车辆具体行驶过程中进行的数据采集,因此,精度也会比较高。
另外,还可以在具体利用上述纵向控制模型进行自动驾驶控制的过程中,获得车辆底盘控制系统在目标速度、角度情况下,响应加速/制动控制部件的量化值信息后,车辆获得的实际加速度信息,并利用所述实际加速度信息对所述车辆纵向运动模型进行校正。也就是说,可以实现实际自动驾驶过程中对模型的校正,使得最终每个车辆的模型能够更准确的反应车辆自身的情况,降低不同车辆之间的轻微不一致性带来的影响,进一步提高精确性。
实施例二
该实施例二是与实施例一相对应的,从自动驾驶计算平台的角度,提供了一种自动驾驶控制方法,参见图3,该方法具体可以包括:
S301:获得车辆纵向运动模型,其中,所述纵向运动模型中包括车辆的加速/制动控制部件的量化值、速度、角度、加速度之间的对应关系;
具体实现时,如实施例一所述,所述纵向运动模型是在控制车辆行驶过程中采集到的数据进行创建的,其中,在采集过程中对车辆的控制包括:在控制车辆获得目标采样角度后,通过逐渐改变加速/制动控制部件的量化值,改变车辆的速度以及加速度。
S302:接收运动控制算法输出的目标加速度信息,并获得车辆的当前速度以及角度信息;
其中,具体的目标加速度信息可以是由计算平台中的路径规划模块、运动控制模块等,根据车辆自身以及周边环境的当前情况计算获得的,也即,需要控制车辆改变为该目标加速度,才能避免与其他车辆发生碰撞,或者,为当前车辆获得更好的行驶空间以及性能,等等。
S303:根据所述纵向运动模型,确定在所述目标加速度信息、当前速度以及角度信息的条件下,对应的加速/制动踏板控制部件的量化值;
其中,所述纵向运动模型中包括在预置的多个目标采样角度条件下,加速/制动踏板开度、速度、加速度之间的对应关系。
但是,由于所述目标采样角度的数量可能是比较有限的,而车辆实际行驶过程中的角度确实多变的,因此,还可以以所述目标采样角度为节点划分为多个角度区间;在当前需要进行控制时,确定所述当前角度所属的角度区间,以及该区间起点对应的第一目标采样角度以及终点对应的第二目标采样角度;并确定所述第一目标采样角度条件下,所述目标加速度信息、当前速度对应的加速/制动踏板的第一开度信息,以及所述第二目标采样角度条件下,所述目标加速度信息、当前速度对应的加速/制动踏板的第二开度信息;然后,可以根据所述第一开度信息、第二开度信息以及预置的线性函数关系信息,确定所述当前角度下,所述目标加速度信息、当前速度对应的加速/制动踏板的第三开度信息。
其中,所述线性函数关系信息是根据以下特性确定:在速度以及加速度固定的情况下,车辆的角度与加速/制动踏板的开度信息之间成线性关系。
S304:根据所述加速/制动控制部件的量化值信息向车辆底盘控制系统发送控制指令。
在向车辆底盘控制系统发送控制指令后,车辆底盘控制系统可以响应该指令,通过改变加速/制动踏板开度信息来改变车辆的加速度。
具体实现时,还可以获得所述车辆底盘控制系统响应所述加速/制动踏板开度信息后,车辆获得的实际加速度信息;利用所述实际加速度信息对所述车辆纵向运动模型进行校正。
关于该实施例二中的未详述部分,可以参见前述实施例一中的记载,这里不再赘述。
实施例三
前述各实施例中,是通过车辆的加速/制动控制部件的量化值信息、速度、角度、加速度之间的对应关系来表达车辆的纵向运动模型,而在实际应用中,纵向运动模型的具体表达方式还可以有多种,因此,本申请实施例还提供了一种自动驾驶控制方法,参见图4,该方法具体可以包括:
S401:获得车辆纵向运动模型;
S402:接收运动控制算法输出的纵向控制目标信息;
其中,纵向控制目标通常也可以是具体的目标加速度信息,或者,还可能包括速度信息等。
S403:根据所述纵向运动模型、所述纵向控制目标信息以及车辆的当前运动状态信息,确定加速/制动控制部件的量化值信息;
当前运动状态信息就可以包括当前的速度、车辆角度等等。
S404:根据所述加速/制动控制部件的量化值信息信息向车辆底盘控制系统发送控制指令。
关于该实施例三中的未详述部分,也可以参见前述实施例一、二中的记载,这里不再赘述。
与实施例一相对应,本申请实施例还提供了一种建立车辆纵向运动模型的装置,参见图5,该装置具体可以包括:
角度控制单元501,用于控制车辆方向盘旋转至目标采样角度;
速度控制单元502,用于在该目标采样角度的条件下,通过控制加速控制部件的量化值将车辆逐渐加速至该目标采样角度条件下的最高安全行车速度,再通过控制制动控制部件的量化值将车辆逐渐减速至停止;
信息采集单元503,用于在所述车辆加速/减速的过程中,按照预置的时间间隔采集并记录加速/制动控制部件的量化值信息与车辆速度、加速度之间的对应关系信息;
模型建立单元504,用于根据多个目标采样角度条件下获得的多组所述对应关系,建立所述车辆纵向运动模型。
具体实现时,该装置还可以包括:
切分单元,用于针对所述多个目标采样角度条件下获得的多组所述对应关系,分别从速度维度上,按照预置的速度间隔进行切分。
滤波单元,用于对所述对应关系中的加速度信息进行滤波处理。
实际加速度获得单元,用于在利用所述纵向运动模型对车辆进行自动驾驶控制的过程中,获得车辆底盘控制系统在目标速度、角度情况下,响应加速/制动控制部件的量化值信息后,车辆获得的实际加速度信息;
校正单元,用于利用所述实际加速度信息对所述车辆纵向运动模型进行校正。
与实施例二相对应,本申请实施例还提供了一种自动驾驶控制装置,参见图6,该装置具体可以包括:
模型获得单元601,用于获得车辆纵向运动模型,其中,所述纵向运动模型中包括车辆的加速/制动控制部件的量化值、速度、角度、加速度之间的对应关系;
信息获得单元602,用于接收运动控制算法输出的目标加速度信息,并获得车辆的当前速度以及角度信息;
量化值信息确定单元603,用于根据所述纵向运动模型,确定在所述目标加速度信 息、当前速度以及角度信息的条件下,对应的加速/制动控制部件的量化值信息;
控制指令发送单元604,用于根据所述加速/制动控制部件的量化值信息向车辆底盘控制系统发送控制指令。
具体实现时,所述纵向运动模型可以是在控制车辆行驶过程中采集到的数据进行创建的,其中,在采集过程中对车辆的控制包括:在控制车辆获得目标采样角度后,通过逐渐改变加速/制动控制部件的量化值信息,改变车辆的速度以及加速度。
其中,所述纵向运动模型中包括在预置的多个目标采样角度条件下,加速/制动控制部件的量化值信息、速度、加速度之间的对应关系。
此时,所述开度信息确定单元具体可以包括:
区间划分子单元,用于以所述目标采样角度为节点划分为多个角度区间;
所属区间确定子单元,用于确定所述当前角度所属的角度区间,以及该区间起点对应的第一目标采样角度以及终点对应的第二目标采样角度;
第一确定子单元,用于确定所述第一目标采样角度条件下,所述目标加速度信息、当前速度对应的加速/制动控制部件的第一量化值信息,以及所述第二目标采样角度条件下,所述目标加速度信息、当前速度对应的加速/制动控制部件的第二量化值信息;
第二确定子单元,用于根据所述第一量化值信息、第二量化值信息以及预置的线性函数关系信息,确定所述当前角度下,所述目标加速度信息、当前速度对应的加速/制动控制部件的第三量化值信息。
其中,所述线性函数关系信息是根据以下特性确定:在速度以及加速度固定的情况下,车辆的角度与加速/制动控制部件的量化值信息之间成线性关系。
具体实现时,该装置还可以包括:
实际加速度信息获得单元,用于获得所述车辆底盘控制系统响应所述加速/制动控制部件的量化值信息后,车辆获得的实际加速度信息;
校正单元,用于利用所述实际加速度信息对所述车辆纵向运动模型进行校正。
与实施例三相对应,本申请实施例还提供了一种自动驾驶控制装置,参见图7,该装置可以包括:
模型获得单元701,用于获得车辆纵向运动模型;
纵向控制目标信息接收单元702,用于接收运动控制算法输出的纵向控制目标信息;
量化值确定单元703,用于根据所述纵向运动模型、所述纵向控制目标信息以及车辆的当前运动状态信息,确定加速/制动控制部件的量化值信息;
控制指令发送单元704,用于根据所述加速/制动控制部件的量化值信息信息向车辆底盘控制系统发送控制指令。
另外,本申请实施例还提供了一种计算机系统,包括:
一个或多个处理器;以及
与所述一个或多个处理器关联的存储器,所述存储器用于存储程序指令,所述程序指令在被所述一个或多个处理器读取执行时,执行如下操作:
获得车辆纵向运动模型,其中,所述纵向运动模型中包括车辆的加速/制动控制部件的量化值、速度、角度、加速度之间的对应关系;
接收运动控制算法输出的目标加速度信息,并获得车辆的当前速度以及角度信息;
根据所述纵向运动模型,确定在所述目标加速度信息、当前速度以及角度信息的条件下,对应的加速/制动控制部件的量化值信息;
根据所述加速/制动控制部件的量化值信息向车辆底盘控制系统发送控制指令。
以及另一种计算机系统,包括:
一个或多个处理器;以及
与所述一个或多个处理器关联的存储器,所述存储器用于存储程序指令,所述程序指令在被所述一个或多个处理器读取执行时,执行如下操作:
控制车辆方向盘旋转至目标采样角度;
在该目标采样角度的条件下,通过控制加速控制部件的量化值将车辆逐渐加速至该目标采样角度条件下的最高安全行车速度,再通过控制制动控制部件的量化值将车辆逐渐减速至停止;
在所述车辆加速/减速的过程中,按照预置的时间间隔采集并记录加速/制动控制部件的量化值与车辆速度、加速度之间的对应关系信息;
根据多个目标采样角度条件下获得的多组所述对应关系,建立所述车辆纵向运动模型。
其中,图8示例性的展示出了计算机系统的架构,具体可以包括处理器810,视频显示适配器811,磁盘驱动器812,输入/输出接口813,网络接口814,以及存储器820。上述处理器810、视频显示适配器811、磁盘驱动器812、输入/输出接口813、网络接口814,与存储器820之间可以通过通信总线830进行通信连接。
其中,处理器810可以采用通用的CPU(Central Processing Unit,中央处理器)、微处理器、应用专用集成电路(Application Specific Integrated Circuit,ASIC)、 或者一个或多个集成电路等方式实现,用于执行相关程序,以实现本申请所提供的技术方案。
存储器820可以采用ROM(Read Only Memory,只读存储器)、RAM(Random Access Memory,随机存取存储器)、静态存储设备,动态存储设备等形式实现。存储器820可以存储用于控制计算机系统800运行的操作系统821,用于控制计算机系统800的低级别操作的基本输入输出系统(BIOS)822。另外,还可以存储网页浏览器823,数据存储管理系统824,以及自动驾驶处理系统825等等。上述自动驾驶处理系统825就可以是本申请实施例中具体实现前述各步骤操作的应用程序。总之,在通过软件或者固件来实现本申请所提供的技术方案时,相关的程序代码保存在存储器820中,并由处理器810来调用执行。
输入/输出接口813用于连接输入/输出模块,以实现信息输入及输出。输入输出/模块可以作为组件配置在设备中(图中未示出),也可以外接于设备以提供相应功能。其中输入设备可以包括键盘、鼠标、触摸屏、麦克风、各类传感器等,输出设备可以包括显示器、扬声器、振动器、指示灯等。
网络接口814用于连接通信模块(图中未示出),以实现本设备与其他设备的通信交互。其中通信模块可以通过有线方式(例如USB、网线等)实现通信,也可以通过无线方式(例如移动网络、WIFI、蓝牙等)实现通信。
总线830包括一通路,在设备的各个组件(例如处理器810、视频显示适配器811、磁盘驱动器812、输入/输出接口813、网络接口814,与存储器820)之间传输信息。
另外,该计算机系统800还可以从虚拟资源对象领取条件信息数据库841中获得具体领取条件的信息,以用于进行条件判断,等等。
需要说明的是,尽管上述设备仅示出了处理器810、视频显示适配器811、磁盘驱动器812、输入/输出接口813、网络接口814,存储器820,总线830等,但是在具体实施过程中,该设备还可以包括实现正常运行所必需的其他组件。此外,本领域的技术人员可以理解的是,上述设备中也可以仅包含实现本申请方案所必需的组件,而不必包含图中所示的全部组件。
通过以上的实施方式的描述可知,本领域的技术人员可以清楚地了解到本申请可借助软件加必需的通用硬件平台的方式来实现。基于这样的理解,本申请的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来,该计算机软件产品可以存储在存储介质中,如ROM/RAM、磁碟、光盘等,包括若干指令用以使得一台计 算机设备(可以是个人计算机,服务器,或者网络设备等)执行本申请各个实施例或者实施例的某些部分所述的方法。
本说明书中的各个实施例均采用递进的方式描述,各个实施例之间相同相似的部分互相参见即可,每个实施例重点说明的都是与其他实施例的不同之处。尤其,对于系统或系统实施例而言,由于其基本相似于方法实施例,所以描述得比较简单,相关之处参见方法实施例的部分说明即可。以上所描述的系统及系统实施例仅仅是示意性的,其中所述作为分离部件说明的单元可以是或者也可以不是物理上分开的,作为单元显示的部件可以是或者也可以不是物理单元,即可以位于一个地方,或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部模块来实现本实施例方案的目的。本领域普通技术人员在不付出创造性劳动的情况下,即可以理解并实施。
以上对本申请所提供的建立车辆纵向运动模型的方法、装置及计算机系统,进行了详细介绍,本文中应用了具体个例对本申请的原理及实施方式进行了阐述,以上实施例的说明只是用于帮助理解本申请的方法及其核心思想;同时,对于本领域的一般技术人员,依据本申请的思想,在具体实施方式及应用范围上均会有改变之处。综上所述,本说明书内容不应理解为对本申请的限制。

Claims (16)

  1. 一种自动驾驶控制方法,其特征在于,包括:
    获得车辆纵向运动模型,其中,所述纵向运动模型中包括车辆的加速/制动控制部件的量化值信息、速度、角度、加速度之间的对应关系;
    接收运动控制算法输出的目标加速度信息,并获得车辆的当前速度以及角度信息;
    根据所述纵向运动模型,确定在所述目标加速度信息、当前速度以及角度信息的条件下,对应的加速/制动控制部件的量化值信息;
    根据所述加速/制动控制部件的量化值信息信息向车辆底盘控制系统发送控制指令。
  2. 根据权利要求1所述的方法,其特征在于,
    所述纵向运动模型是在控制车辆行驶过程中采集到的数据进行创建的,其中,在采集过程中对车辆的控制包括:在控制车辆获得目标采样角度后,通过逐渐改变加速/制动控制部件的量化值信息,改变车辆的速度以及加速度。
  3. 根据权利要求2所述的方法,其特征在于,
    所述纵向运动模型中包括在预置的多个目标采样角度条件下,加速/制动控制部件的量化值信息、速度、加速度之间的对应关系。
  4. 根据权利要求3所述的方法,其特征在于,
    所述根据所述纵向运动模型,确定在所述目标加速度信息、当前速度以及角度信息的条件下,对应的加速/制动控制部件的量化值信息,包括:
    以所述目标采样角度为节点划分为多个角度区间;
    确定所述当前角度所属的角度区间,以及该区间起点对应的第一目标采样角度以及终点对应的第二目标采样角度;
    确定所述第一目标采样角度条件下,所述目标加速度信息、当前速度对应的加速/制动控制部件的第一量化值信息,以及所述第二目标采样角度条件下,所述目标加速度信息、当前速度对应的加速/制动控制部件的第二量化值信息;
    根据所述第一量化值信息、第二量化值信息以及预置的线性函数关系信息,确定所述当前角度下,所述目标加速度信息、当前速度对应的加速/制动控制部件的第三量化值信息。
  5. 根据权利要求4所述的方法,其特征在于,
    所述线性函数关系信息是根据以下特性确定:在速度以及加速度固定的情况下,车辆的角度与加速/制动控制部件的量化值信息之间成线性关系。
  6. 根据权利要求1所述的方法,其特征在于,还包括:
    获得所述车辆底盘控制系统响应所述加速/制动控制部件的量化值信息后,车辆获得的实际加速度信息;
    利用所述实际加速度信息对所述车辆纵向运动模型进行校正。
  7. 一种建立车辆纵向运动模型的方法,其特征在于,包括:
    控制车辆方向盘旋转至目标采样角度;
    在该目标采样角度的条件下,通过控制加速控制部件的量化值将车辆逐渐加速至该目标采样角度条件下的最高安全行车速度,再通过控制制动控制部件的量化值将车辆逐渐减速至停止;
    在所述车辆加速/减速的过程中,按照预置的时间间隔采集并记录加速/制动控制部件的量化值与车辆速度、加速度之间的对应关系信息;
    根据多个目标采样角度条件下获得的多组所述对应关系,建立所述车辆纵向运动模型。
  8. 根据权利要求7所述的方法,其特征在于,还包括:
    针对所述多个目标采样角度条件下获得的多组所述对应关系,分别从速度维度上,按照预置的速度间隔进行切分。
  9. 根据权利要求7所述的方法,其特征在于,还包括:
    对所述对应关系中的加速度信息进行滤波处理。
  10. 根据权利要求7所述的方法,其特征在于,还包括:
    在利用所述纵向运动模型对车辆进行自动驾驶控制的过程中,获得车辆底盘控制系统在目标速度、角度情况下,响应加速/制动控制部件的量化值信息后,车辆获得的实际加速度信息;
    利用所述实际加速度信息对所述车辆纵向运动模型进行校正。
  11. 一种自动驾驶控制方法,其特征在于,包括:
    获得车辆纵向运动模型;
    接收运动控制算法输出的纵向控制目标信息;
    根据所述纵向运动模型、所述纵向控制目标信息以及车辆的当前运动状态信息,确定加速/制动控制部件的量化值信息;
    根据所述加速/制动控制部件的量化值信息信息向车辆底盘控制系统发送控制指令。
  12. 一种自动驾驶控制装置,其特征在于,包括:
    模型获得单元,用于获得车辆纵向运动模型,其中,所述纵向运动模型中包括车辆的加速/制动控制部件的量化值、速度、角度、加速度之间的对应关系;
    信息获得单元,用于接收运动控制算法输出的目标加速度信息,并获得车辆的当前速度以及角度信息;
    量化值信息确定单元,用于根据所述纵向运动模型,确定在所述目标加速度信息、当前速度以及角度信息的条件下,对应的加速/制动控制部件的量化值信息;
    控制指令发送单元,用于根据所述加速/制动踏板控制部件的量化值信息向车辆底盘控制系统发送控制指令。
  13. 一种建立车辆纵向运动模型的装置,其特征在于,包括:
    角度控制单元,用于控制车辆方向盘旋转至目标采样角度;
    踏板开度控制单元,用于在该目标采样角度的条件下,通过控制加速控制部件的量化值将车辆逐渐加速至该目标采样角度条件下的最高安全行车速度,再通过控制制动控制部件的量化值将车辆逐渐减速至停止;
    信息采集单元,用于在所述车辆加速/减速的过程中,按照预置的时间间隔采集并记录加速/制动控制部件的量化值与车辆速度、加速度之间的对应关系信息;
    模型建立单元,用于根据多个目标采样角度条件下获得的多组所述对应关系,建立所述车辆纵向运动模型。
  14. 一种自动驾驶控制装置,其特征在于,包括:
    模型获得单元,用于获得车辆纵向运动模型;
    纵向控制目标信息接收单元,用于接收运动控制算法输出的纵向控制目标信息;
    量化值确定单元,用于根据所述纵向运动模型、所述纵向控制目标信息以及车辆的当前运动状态信息,确定加速/制动控制部件的量化值信息;
    指令发送单元,用于根据所述加速/制动控制部件的量化值信息信息向车辆底盘控制系统发送控制指令。
  15. 一种计算机系统,其特征在于,包括:
    一个或多个处理器;以及
    与所述一个或多个处理器关联的存储器,所述存储器用于存储程序指令,所述程序指令在被所述一个或多个处理器读取执行时,执行如下操作:
    获得车辆纵向运动模型,其中,所述纵向运动模型中包括车辆的加速/制动控制部件的量化值、速度、角度、加速度之间的对应关系;
    接收运动控制算法输出的目标加速度信息,并获得车辆的当前速度以及角度信息;
    根据所述纵向运动模型,确定在所述目标加速度信息、当前速度以及角度信息的条件下,对应的加速/制动控制部件的量化值信息;
    根据所述加速/制动控制部件的量化值信息向车辆底盘控制系统发送控制指令。
  16. 一种计算机系统,其特征在于,包括:
    一个或多个处理器;以及
    与所述一个或多个处理器关联的存储器,所述存储器用于存储程序指令,所述程序指令在被所述一个或多个处理器读取执行时,执行如下操作:
    控制车辆方向盘旋转至目标采样角度;
    在该目标采样角度的条件下,通过控制加速控制部件的量化值将车辆逐渐加速至该目标采样角度条件下的最高安全行车速度,再通过控制制动控制部件的量化值将车辆逐渐减速至停止;
    在所述车辆加速/减速的过程中,按照预置的时间间隔采集并记录加速/制动控制部件的量化值与车辆速度、加速度之间的对应关系信息;
    根据多个目标采样角度条件下获得的多组所述对应关系,建立所述车辆纵向运动模型。
PCT/CN2020/076530 2019-03-05 2020-02-25 建立车辆纵向运动模型的方法、装置及计算机系统 WO2020177571A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910162589.8A CN111661060B (zh) 2019-03-05 2019-03-05 建立车辆纵向运动模型的方法、装置及计算机系统
CN201910162589.8 2019-03-05

Publications (1)

Publication Number Publication Date
WO2020177571A1 true WO2020177571A1 (zh) 2020-09-10

Family

ID=72338117

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/076530 WO2020177571A1 (zh) 2019-03-05 2020-02-25 建立车辆纵向运动模型的方法、装置及计算机系统

Country Status (2)

Country Link
CN (1) CN111661060B (zh)
WO (1) WO2020177571A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115793675A (zh) * 2023-02-09 2023-03-14 陕西欧卡电子智能科技有限公司 无人船动力系统的耦合标定方法、装置及计算机设备
CN116424370A (zh) * 2023-06-14 2023-07-14 苏州上善知源汽车电子有限公司 一种车辆控制方法、装置、电子设备及存储介质

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112379668A (zh) * 2020-10-30 2021-02-19 深圳元戎启行科技有限公司 车辆控制数据标定方法、装置、计算机设备和存储介质
CN115056797A (zh) * 2021-03-08 2022-09-16 长沙智能驾驶研究院有限公司 车辆控制方法、装置、设备及计算机存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105752084A (zh) * 2014-12-17 2016-07-13 上汽通用五菱汽车股份有限公司 一种新能源汽车能量优化方法
CN107117167A (zh) * 2017-04-24 2017-09-01 南京航空航天大学 具有多种避撞模式的汽车差速转向系统及其控制方法
CN108312966A (zh) * 2018-02-26 2018-07-24 江苏裕兰信息科技有限公司 一种包含汽车底部影像的全景环视系统及其实现方法
CN108717268A (zh) * 2018-06-22 2018-10-30 南京航空航天大学 基于最优控制与安全距离的自动驾驶最速操纵控制系统及其控制方法
US20180345972A1 (en) * 2017-05-31 2018-12-06 NextEv USA, Inc. Utilization of Smoothing Functions for Acceleration and Deceleration Profile Generation

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE19632337C2 (de) * 1996-08-10 2000-12-14 Daimler Chrysler Ag Verfahren und Einrichtung zur Regelung der Längsdynamik eines Kraftfahrzeuges
JP5780996B2 (ja) * 2012-03-28 2015-09-16 日立オートモティブシステムズ株式会社 車両走行制御装置
CN102795225B (zh) * 2012-09-10 2015-01-21 北京理工大学 利用驾驶员侧纵向控制模型检测驾驶员干扰状态的方法
JP2017144776A (ja) * 2016-02-15 2017-08-24 株式会社Subaru 車両の走行制御装置
CN108427417B (zh) * 2018-03-30 2020-11-24 北京图森智途科技有限公司 自动驾驶控制系统及方法、计算机服务器和自动驾驶车辆

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105752084A (zh) * 2014-12-17 2016-07-13 上汽通用五菱汽车股份有限公司 一种新能源汽车能量优化方法
CN107117167A (zh) * 2017-04-24 2017-09-01 南京航空航天大学 具有多种避撞模式的汽车差速转向系统及其控制方法
US20180345972A1 (en) * 2017-05-31 2018-12-06 NextEv USA, Inc. Utilization of Smoothing Functions for Acceleration and Deceleration Profile Generation
CN108312966A (zh) * 2018-02-26 2018-07-24 江苏裕兰信息科技有限公司 一种包含汽车底部影像的全景环视系统及其实现方法
CN108717268A (zh) * 2018-06-22 2018-10-30 南京航空航天大学 基于最优控制与安全距离的自动驾驶最速操纵控制系统及其控制方法

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115793675A (zh) * 2023-02-09 2023-03-14 陕西欧卡电子智能科技有限公司 无人船动力系统的耦合标定方法、装置及计算机设备
CN116424370A (zh) * 2023-06-14 2023-07-14 苏州上善知源汽车电子有限公司 一种车辆控制方法、装置、电子设备及存储介质
CN116424370B (zh) * 2023-06-14 2023-09-15 苏州上善知源汽车电子有限公司 一种车辆控制方法、装置、电子设备及存储介质

Also Published As

Publication number Publication date
CN111661060A (zh) 2020-09-15
CN111661060B (zh) 2022-06-21

Similar Documents

Publication Publication Date Title
WO2020177571A1 (zh) 建立车辆纵向运动模型的方法、装置及计算机系统
EP3617827A2 (en) Vehicle controlling method and apparatus, computer device, and storage medium
US10053091B2 (en) Spring system-based change lane approach for autonomous vehicles
CN111665738A (zh) 在环仿真系统及其中的信息处理方法及装置
WO2022247203A1 (zh) 自动驾驶车辆的控制方法、装置、设备以及存储介质
US20190339707A1 (en) Automobile Image Processing Method and Apparatus, and Readable Storage Medium
GB2543921A (en) Active vehicle suspension
CN112026769A (zh) 辅助驾驶的纵向控制方法、装置、系统及存储介质
WO2023051224A1 (zh) 自动驾驶车辆的纵向控制方法、装置、设备及介质
CN111098842B (zh) 车辆速度控制方法及相关设备
EP4074569A1 (en) Method for determining automatic driving feature, apparatus, device, medium and program product
CN113442854B (zh) 数据处理方法、装置、电子设备和存储介质
CN115503632A (zh) 无人驾驶矿用车辆制动距离获取方法、装置、电子设备及存储介质
CN113848899A (zh) 车辆横向控制方法、装置、设备及存储介质
WO2024051024A1 (zh) 车辆避障方法、装置、设备及存储介质
CN110171404A (zh) 行车安全距离控制方法、系统、设备及存储介质
CN116088538A (zh) 车辆轨迹信息生成方法、装置、设备和计算机可读介质
CN110834612A (zh) 冗余制动控制方法、装置、系统、车辆和存储介质
CN114419758B (zh) 跟车距离的计算方法、装置、车辆及存储介质
CN115372020A (zh) 自动驾驶车辆测试方法、装置、电子设备和介质
CN115309085A (zh) 无人驾驶双通讯冗余控制方法、系统、电子设备及介质
CN113799772B (zh) 车辆的控制方法、装置以及控制系统
JP2021156889A (ja) 車両体感測定評価方法、装置、電子機器、記憶媒体及びコンピュータプログラム
CN112230659A (zh) 精准规划运动轨迹的方法、智能控制设备及自动驾驶车辆
CN116729336B (zh) 一种制动时间的确定方法、装置、设备和介质

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: 20765608

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20765608

Country of ref document: EP

Kind code of ref document: A1