Environment information prediction, control method and device of autonomous vehicle and vehicle

Abstract

The present application relates to the technical field of automatic driving, and particularly provides an automatic driving vehicle environment information prediction and control method, device and vehicle, aiming to solve the problem of how to combine various information to construct an environment model of the automatic driving vehicle, so as to fully consider the environment information and improve the robustness of the model. To this end, the present application constructs a state transition model of Kalman filtering equation according to the ego motion information of the automatic driving vehicle, further obtains a predicted value of the position relationship of the environment target at the current time, updates the predicted value by applying the Kalman filtering equation based on the predicted value and the current time perception measurement result, obtains the final prediction result, and the perception measurement result is obtained through the ego motion system, the vehicle-end perception device and the non-vehicle-end perception device, which can comprehensively predict the motion trend of the environment target, has higher robustness, and can provide more comprehensive, stable and reliable control input for the control process of the automatic driving vehicle.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving technology, specifically providing an autonomous vehicle environmental information prediction and control method, device, and vehicle. Background Technology

[0002] With the development of autonomous driving technology, future autonomous vehicles will be equipped with an increasing number of sensors, such as radar, lidar, high-precision maps, and high-precision inertial measurement units (IMUs). Thanks to breakthroughs in image processing, target fusion, and localization technologies, acquiring multi-dimensional perception information is now readily available. Utilizing this rich information provides the conditions for obtaining more accurate and stable vehicle environment models. Furthermore, based on the construction of these environment models, a foundation is laid for predicting the behavior of surrounding traffic participants on the road, thus providing a more stable and reliable guarantee for autonomous vehicles operating on public roads.

[0003] Vehicle environment models play a crucial role in the behavior planning and control of autonomous vehicles. They can predict the driving trajectories of the vehicle itself and surrounding vehicles, thereby acquiring information on road features and obstacles along the driving trajectory and predicting the trajectory. This provides input for the vehicle's decision-making and movement in the next stage, ultimately enabling the control processes of autonomous vehicles, such as steering control, drive control, and braking control.

[0004] Currently, vehicle environment models lack a standardized form, and most do not consider the prediction of traffic participant motion states. Early models relied on single pieces of information for prediction, such as vehicle speed and yaw rate. These are steady-state estimation methods based on the current vehicle state and do not effectively represent future vehicle motion trends. While perception information based on road attributes can represent road information in a certain time and space, the robustness of the acquired environment models is limited by obstacles and weather. Furthermore, applying multiple types of information to acquire environment models is constrained by the lack of uniformity in the acquisition of different types of perception information, making it difficult to fuse and use multiple input information.

[0005] Accordingly, there is a need in this field for a new scheme for constructing environmental models for autonomous vehicles to address the above problems. Summary of the Invention

[0006] To overcome the above-mentioned shortcomings, this invention is proposed to provide a solution, or at least a partial solution, to the problem of how to combine multiple types of information to construct an environmental model for autonomous vehicles, thereby fully considering environmental information and improving the robustness of the model.

[0007] In a first aspect, the present invention provides a method for predicting environmental information for an autonomous vehicle, the method comprising:

[0008] Based on the vehicle motion information of the autonomous vehicle, a state transition model of the Kalman filter equation is constructed;

[0009] Based on the state transition model, the predicted values ​​of the positional relationships of environmental targets in the surrounding environment of the autonomous vehicle at the current moment are obtained; wherein, the positional relationship is the relationship between the environmental targets and the position of the autonomous vehicle.

[0010] Based on the perception measurement results of the positional relationship of the environmental targets at the current moment obtained by the autonomous vehicle and the predicted value, the predicted value is updated based on the Kalman filter equation to obtain the final predicted result of the positional relationship of the environmental targets at the current moment.

[0011] The perception measurement results are obtained through the autonomous vehicle's motion system and / or vehicle-side perception devices and / or non-vehicle-side perception devices.

[0012] In one technical solution of the above-mentioned method for predicting environmental information of autonomous vehicles, the environmental target includes the predicted trajectory point of the autonomous vehicle within a preset aiming time, and the positional relationship includes the longitudinal distance and orientation angle of the predicted trajectory point relative to the autonomous vehicle.

[0013] The method further includes:

[0014] Based on the autonomous vehicle's motion system, the vehicle speed and yaw angle of the autonomous vehicle are obtained;

[0015] The autonomous vehicle's driving radius is obtained based on the vehicle speed and the yaw angle.

[0016] Based on the autonomous vehicle's driving radius, the vehicle's driving trajectory is obtained;

[0017] Based on the vehicle's driving trajectory, the longitudinal distance and orientation angle of the predicted trajectory points before and after the vehicle's driving trajectory are obtained within the preceding and following aiming time, and are used as the perception measurement results of the predicted trajectory points.

[0018] In one technical solution of the above-mentioned method for predicting environmental information of autonomous vehicles, the step of "obtaining the autonomous vehicle's driving radius based on the vehicle speed and yaw angle" includes:

[0019] Based on the assumption that the autonomous vehicle is moving in a fixed circle, the vehicle's driving radius at each moment is obtained according to the vehicle speed and yaw angle.

[0020] In one technical solution of the above-mentioned method for predicting environmental information of autonomous vehicles, the environmental target includes road attributes, and the positional relationship includes the longitudinal distance and orientation angle of the road attributes relative to the autonomous vehicle.

[0021] The method further includes:

[0022] Based on the vehicle-mounted perception devices of the autonomous vehicle, the perception data of the road attributes are obtained;

[0023] The perceived data is discretized based on multiple discrete points of the road attributes;

[0024] Based on the plurality of discrete points, the longitudinal distance and orientation angle of the discrete points relative to the autonomous vehicle are obtained as the perception measurement results of the road attributes.

[0025] In one technical solution of the above-mentioned method for predicting environmental information of autonomous vehicles, the environmental target includes traffic participants, and the positional relationship includes the longitudinal distance and orientation angle of the traffic participants relative to the autonomous vehicle;

[0026] The method further includes:

[0027] The perception information of the traffic participants is obtained based on the vehicle-mounted perception device of the autonomous vehicle.

[0028] Based on the perception information of traffic participants, the displacement of the traffic participants in two frames of perception information at adjacent sampling times is obtained;

[0029] Based on the current position and displacement of the autonomous vehicle, the longitudinal distance and orientation angle of the traffic participant relative to the autonomous vehicle are obtained as the perception measurement results of the traffic participant.

[0030] In one technical solution of the above-mentioned method for predicting environmental information for autonomous vehicles, the traffic participants include non-lane-changing vehicles;

[0031] The method further includes:

[0032] The speeds of surrounding vehicles and the angles between the current and previous times of the surrounding vehicles are obtained for the autonomous vehicle.

[0033] The surrounding vehicles whose speed is greater than a preset speed and whose angle is less than a preset angle are considered as the non-lane-changing vehicles.

[0034] In one technical solution of the above-mentioned method for predicting environmental information of autonomous vehicles, the step of "updating the predicted value based on the Kalman filter equation according to the perception measurement results of the autonomous vehicle at the current moment and the predicted value" includes:

[0035] Based on the multiple orientation angles of the non-lane-changing vehicles, obtain the mean and variance of the orientation angles;

[0036] Based on the mean and the preset mean, determine the valid non-lane-changing vehicles;

[0037] The predicted value is updated based on the longitudinal distance and orientation angle of the effective non-lane-changing vehicles.

[0038] In one technical solution of the above-mentioned method for predicting environmental information of autonomous vehicles, the step of "updating the predicted value based on the longitudinal distance and orientation angle of the effective non-lane-changing vehicle" includes:

[0039] The variance of the orientation angle of the non-lane-changing vehicles is used as noise to update the Kalman filter equation;

[0040] The predicted values ​​are updated based on the noise, as well as the longitudinal distance and orientation angle of the effective non-lane-changing vehicles.

[0041] In one technical solution of the above-mentioned method for predicting environmental information of autonomous vehicles, the environmental target includes the lane path, and the positional relationship includes the longitudinal distance and orientation angle between the lane path and the autonomous vehicle.

[0042] The method further includes:

[0043] The driving path of the autonomous vehicle is obtained based on the non-vehicle-side perception devices of the autonomous vehicle.

[0044] Based on the current location of the autonomous vehicle and the driving path, obtain the lane path discrete points of the lane where the autonomous vehicle is currently located and the surrounding lanes.

[0045] The longitudinal distance and orientation angle of the discrete points of the lane path relative to the autonomous vehicle are used as the perception measurement results of the lane path.

[0046] In one technical solution of the above-mentioned method for predicting environmental information of autonomous vehicles, the step of "obtaining the lane path discrete points of the lane where the autonomous vehicle is currently located and the surrounding lanes based on the current position of the autonomous vehicle and the driving path" includes:

[0047] When roads merge, the discrete points of the lane path behind the autonomous vehicle are obtained based on the current location;

[0048] The step of “using the longitudinal distance and orientation angle of the discrete points of the lane path relative to the autonomous vehicle as the perception measurement result of the lane path” includes:

[0049] The longitudinal distance and orientation angle of the discrete points of the rear lane path relative to the autonomous vehicle are used as the perception measurement results of the lane path.

[0050] In one technical solution of the above-mentioned method for predicting environmental information of autonomous vehicles, before the step of "obtaining the predicted values ​​of the positional relationships of environmental targets in the surrounding environment of the autonomous vehicle at the current moment according to the state transition model", the method further includes:

[0051] Initialize the perceived measurement information of the environmental target;

[0052] Based on the initialization results, the effectiveness of the sensing measurement information acquisition device is determined.

[0053] In one technical solution of the above-mentioned method for predicting environmental information of autonomous vehicles, the step of "determining the effectiveness of the device for acquiring the perception measurement information" includes:

[0054] When the acquisition device is the vehicle-side perception device of the autonomous vehicle, the comprehensive confidence level of the vehicle-side perception device is obtained based on the initial confidence level of the vehicle-side perception device and the environmental information.

[0055] Based on the comprehensive confidence level, the effectiveness of the vehicle-mounted sensing device is determined; and / or,

[0056] When the acquisition device is a map matching system, the validity is determined based on the communication between different maps in the map matching system, whether the positioning system of the autonomous vehicle is working properly, and whether the current road is covered by a high-precision map.

[0057] In one technical solution of the above-mentioned method for predicting environmental information of autonomous vehicles, before the step of "updating the predicted value based on the Kalman filter equation according to the perception measurement results of the positional relationship of the environmental target at the current time and the predicted value obtained by the autonomous vehicle, and obtaining the final prediction result of the positional relationship of the environmental target at the current time", the method includes:

[0058] Obtain the confidence level of the perceived measurement results;

[0059] Based on the confidence level, the predicted value is selectively updated using the perceived measurement results to obtain the final prediction result.

[0060] In one technical solution of the above-mentioned method for predicting environmental information of autonomous vehicles, the environmental targets include moving targets and stationary targets. The step of "updating the predicted values ​​based on the Kalman filter equation according to the perception measurement results of the positional relationship of the environmental targets at the current moment obtained by the autonomous vehicle and the predicted values, and obtaining the final predicted results of the positional relationship of the environmental targets at the current moment" includes:

[0061] The predicted value is updated based on the perception measurement results of the moving target to obtain the updated predicted value of the moving target;

[0062] The final prediction result is obtained based on the updated predicted value of the moving target and the positional relationship between the stationary target and the autonomous vehicle.

[0063] In a second aspect, the present invention provides a control method for an autonomous vehicle, the method comprising:

[0064] According to any one of the above-mentioned environmental information prediction methods for autonomous vehicles, the final prediction result of environmental targets in the surrounding environment of the autonomous vehicle is obtained.

[0065] The autonomous vehicle is controlled based on the final prediction result.

[0066] In one technical solution of the above-mentioned control method for autonomous vehicles, the step of "controlling the autonomous vehicle according to the final prediction result" includes:

[0067] The environmental target is encoded based on the final prediction result;

[0068] The autonomous vehicle is controlled based on the encoding result.

[0069] In one technical solution of the above-mentioned control method for autonomous vehicles, the step of "the environmental target being the vehicles surrounding the autonomous vehicle; and encoding the environmental target based on the final prediction result" includes:

[0070] Based on the final prediction results of the surrounding vehicles, obtain the orientation angle of the surrounding vehicles;

[0071] The surrounding vehicles are encoded based on the orientation angle and a preset included angle threshold.

[0072] In a third aspect, a control device is provided, comprising a processor and a storage device, the storage device being adapted to store a plurality of program codes, the program codes being adapted to be loaded and run by the processor to execute the environmental information prediction method for an autonomous vehicle as described in any of the above-described technical solutions of the autonomous vehicle environmental information prediction method or the control method for an autonomous vehicle as described in any of the above-described technical solutions of the autonomous vehicle control method.

[0073] In a fourth aspect, a computer-readable storage medium is provided, wherein a plurality of program codes are stored therein, the program codes being adapted to be loaded and run by a processor to perform the environmental information prediction method for an autonomous vehicle as described in any of the above-described technical solutions of the autonomous vehicle environmental information prediction method or the control method for an autonomous vehicle as described in any of the above-described technical solutions of the autonomous vehicle control method.

[0074] In a fifth aspect, a vehicle is provided, the vehicle including the control device described above.

[0075] The above-described technical solutions of the present invention have at least one or more of the following beneficial effects:

[0076] In implementing the technical solution of this invention, a state transition model of the Kalman filter equation is constructed based on the vehicle's motion information. The predicted positional relationships of environmental targets in the vehicle's surroundings at the current moment are obtained using the state transition model. Based on the predicted values ​​and the perception measurement results of the current positional relationships, the Kalman filter equation is applied to update the predicted values, obtaining the final predicted result of the positional relationships of the environmental targets at the current moment. The perception measurement results can be obtained through the vehicle's motion system, vehicle-side perception devices, and non-vehicle-side perception devices. Through this configuration, this invention predicts the positional relationships between environmental targets and the autonomous vehicle based on the Kalman filter model, and updates the predicted values ​​of the environmental targets' positional relationships based on the multi-dimensional perception measurement results of the autonomous vehicle. This enables a more comprehensive prediction of the motion trends of environmental targets in the vehicle's surroundings, exhibiting higher robustness. Simultaneously, it allows control of the autonomous vehicle based on the final predicted results of the environmental information, providing a more comprehensive, stable, and reliable control input for the autonomous vehicle's control process, achieving more effective control.

[0077] Solution 1. A method for predicting environmental information of an autonomous vehicle, characterized in that the method includes:

[0078] Based on the vehicle motion information of the autonomous vehicle, a state transition model of the Kalman filter equation is constructed;

[0079] Based on the state transition model, the predicted values ​​of the positional relationships of environmental targets in the surrounding environment of the autonomous vehicle at the current moment are obtained; wherein, the positional relationship is the relationship between the environmental targets and the position of the autonomous vehicle.

[0080] Based on the perception measurement results of the positional relationship of the environmental targets at the current moment obtained by the autonomous vehicle and the predicted value, the predicted value is updated based on the Kalman filter equation to obtain the final predicted result of the positional relationship of the environmental targets at the current moment.

[0081] The perception measurement results are obtained through the autonomous vehicle's motion system and / or vehicle-side perception devices and / or non-vehicle-side perception devices.

[0082] Scheme 2. The environmental information prediction method for autonomous vehicles according to Scheme 1, characterized in that the environmental target includes the predicted trajectory points of the autonomous vehicle within a preset aiming time, and the positional relationship includes the longitudinal distance and orientation angle of the predicted trajectory points relative to the autonomous vehicle;

[0083] The method further includes:

[0084] Based on the autonomous vehicle's motion system, the vehicle speed and yaw angle of the autonomous vehicle are obtained;

[0085] The autonomous vehicle's driving radius is obtained based on the vehicle speed and the yaw angle.

[0086] Based on the autonomous vehicle's driving radius, the vehicle's driving trajectory is obtained;

[0087] Based on the vehicle's driving trajectory, the longitudinal distance and orientation angle of the predicted trajectory points before and after the vehicle's driving trajectory are obtained within the preceding and following aiming time, and are used as the perception measurement results of the predicted trajectory points.

[0088] Solution 3. The environmental information prediction method for autonomous vehicles according to Solution 2, characterized in that the step of "obtaining the autonomous vehicle's driving radius based on the vehicle speed and yaw angle" includes:

[0089] Based on the assumption that the autonomous vehicle is moving in a fixed circle, the vehicle's driving radius at each moment is obtained according to the vehicle speed and yaw angle.

[0090] Solution 4. The environmental information prediction method for autonomous vehicles according to Solution 1, characterized in that the environmental target includes road attributes, and the positional relationship includes the longitudinal distance and orientation angle of the road attributes relative to the autonomous vehicle;

[0091] The method further includes:

[0092] Based on the vehicle-mounted perception devices of the autonomous vehicle, the perception data of the road attributes are obtained;

[0093] The perceived data is discretized based on multiple discrete points of the road attributes;

[0094] Based on the plurality of discrete points, the longitudinal distance and orientation angle of the discrete points relative to the autonomous vehicle are obtained as the perception measurement results of the road attributes.

[0095] Scheme 5. The environmental information prediction method for autonomous vehicles according to Scheme 1, characterized in that the environmental target includes traffic participants, and the positional relationship includes the longitudinal distance and orientation angle of the traffic participants relative to the autonomous vehicle;

[0096] The method further includes:

[0097] The perception information of the traffic participants is obtained based on the vehicle-mounted perception device of the autonomous vehicle.

[0098] Based on the perception information of traffic participants, the displacement of the traffic participants in two frames of perception information at adjacent sampling times is obtained;

[0099] Based on the current position and displacement of the autonomous vehicle, the longitudinal distance and orientation angle of the traffic participant relative to the autonomous vehicle are obtained as the perception measurement results of the traffic participant.

[0100] Scheme 6. The environmental information prediction method for autonomous vehicles according to Scheme 5, wherein the traffic participants include non-lane-changing vehicles;

[0101] The method further includes:

[0102] The speeds of surrounding vehicles and the angles between the current and previous times of the surrounding vehicles are obtained for the autonomous vehicle.

[0103] The surrounding vehicles whose speed is greater than a preset speed and whose angle is less than a preset angle are considered as the non-lane-changing vehicles.

[0104] Solution 7. The environmental information prediction method for autonomous vehicles according to Solution 6, characterized in that the step of "updating the predicted value based on the Kalman filter equation according to the perception measurement results of the autonomous vehicle at the current moment and the predicted value" includes:

[0105] Based on the multiple orientation angles of the non-lane-changing vehicles, obtain the mean and variance of the orientation angles;

[0106] Based on the mean and the preset mean, determine the valid non-lane-changing vehicles;

[0107] The predicted value is updated based on the longitudinal distance and orientation angle of the effective non-lane-changing vehicles.

[0108] Solution 8. The environmental information prediction method for autonomous vehicles according to Solution 7, characterized in that the step of "updating the predicted value based on the longitudinal distance and orientation angle of the effective non-lane-changing vehicle" includes:

[0109] The variance of the orientation angle of the non-lane-changing vehicles is used as noise to update the Kalman filter equation;

[0110] The predicted values ​​are updated based on the noise, as well as the longitudinal distance and orientation angle of the effective non-lane-changing vehicles.

[0111] Scheme 9. The environmental information prediction method for autonomous vehicles according to Scheme 1, characterized in that the environmental target includes the lane path, and the positional relationship includes the longitudinal distance and orientation angle between the lane path and the autonomous vehicle;

[0112] The method further includes:

[0113] The driving path of the autonomous vehicle is obtained based on the non-vehicle-side perception devices of the autonomous vehicle.

[0114] Based on the current location of the autonomous vehicle and the driving path, obtain the lane path discrete points of the lane where the autonomous vehicle is currently located and the surrounding lanes.

[0115] The longitudinal distance and orientation angle of the discrete points of the lane path relative to the autonomous vehicle are used as the perception measurement results of the lane path.

[0116] Solution 10. The environmental information prediction method for autonomous vehicles according to Solution 9, characterized in that,

[0117] The step of "obtaining the lane path discrete points of the lane where the autonomous vehicle is currently located and the surrounding lanes based on the current position of the autonomous vehicle and the driving path" includes:

[0118] When roads merge, the discrete points of the lane path behind the autonomous vehicle are obtained based on the current location;

[0119] The step of “using the longitudinal distance and orientation angle of the discrete points of the lane path relative to the autonomous vehicle as the perception measurement result of the lane path” includes:

[0120] The longitudinal distance and orientation angle of the discrete points of the rear lane path relative to the autonomous vehicle are used as the perception measurement results of the lane path.

[0121] Solution 11. The environmental information prediction method for autonomous vehicles according to Solution 1, characterized in that, before the step of "obtaining the predicted values ​​of the positional relationships of environmental targets in the surrounding environment of the autonomous vehicle at the current moment according to the state transition model", the method further includes:

[0122] Initialize the perceived measurement information of the environmental target;

[0123] Based on the initialization results, the effectiveness of the sensing measurement information acquisition device is determined.

[0124] Solution 12. The environmental information prediction method for autonomous vehicles according to Solution 11, characterized in that the step of "determining the validity of the perception measurement information acquisition device" includes:

[0125] When the acquisition device is the vehicle-side perception device of the autonomous vehicle, the comprehensive confidence level of the vehicle-side perception device is obtained based on the initial confidence level of the vehicle-side perception device and the environmental information.

[0126] Based on the comprehensive confidence level, the effectiveness of the vehicle-mounted sensing device is determined; and / or,

[0127] When the acquisition device is a map matching system, the validity is determined based on the communication between different maps in the map matching system, whether the positioning system of the autonomous vehicle is working properly, and whether the current road is covered by a high-precision map.

[0128] Solution 13. The environmental information prediction method for autonomous vehicles according to Solution 1, characterized in that, before the step of "updating the predicted value based on the Kalman filter equation according to the perception measurement results of the positional relationship of the environmental target at the current time and the predicted value obtained by the autonomous vehicle, and obtaining the final prediction result of the positional relationship of the environmental target at the current time", the method includes:

[0129] Obtain the confidence level of the perceived measurement results;

[0130] Based on the confidence level, the predicted value is selectively updated using the perceived measurement results to obtain the final prediction result.

[0131] Solution 14. The environmental information prediction method for autonomous vehicles according to Solution 1, characterized in that the environmental targets include moving targets and stationary targets, and the step of "updating the predicted values ​​based on the Kalman filter equation according to the perception measurement results of the positional relationship of the environmental targets at the current time obtained by the autonomous vehicle and the predicted values, and obtaining the final prediction result of the positional relationship of the environmental targets at the current time" includes:

[0132] The predicted value is updated based on the perception measurement results of the moving target to obtain the updated predicted value of the moving target;

[0133] The final prediction result is obtained based on the updated predicted value of the moving target and the positional relationship between the stationary target and the autonomous vehicle.

[0134] Solution 15. A control method for an autonomous vehicle, characterized in that the method includes:

[0135] According to any one of the schemes 1 to 14, the environmental information prediction method for autonomous vehicles is used to obtain the final prediction result of environmental targets in the surrounding environment of the autonomous vehicle.

[0136] The autonomous vehicle is controlled based on the final prediction result.

[0137] Solution 16. The control method for an autonomous vehicle according to Solution 15, characterized in that the step of "controlling the autonomous vehicle according to the final prediction result" includes:

[0138] The environmental target is encoded based on the final prediction result;

[0139] The autonomous vehicle is controlled based on the encoding result.

[0140] Solution 17. The control method for an autonomous vehicle according to Solution 16, characterized in that the environmental target is the surrounding vehicles of the autonomous vehicle; the step of "encoding the environmental target according to the final prediction result" includes:

[0141] Based on the final prediction results of the surrounding vehicles, obtain the orientation angle of the surrounding vehicles;

[0142] The surrounding vehicles are encoded based on the orientation angle and a preset included angle threshold.

[0143] Scheme 18. A control device comprising at least one processor and at least one storage device, the storage device being adapted to store a plurality of program codes, characterized in that the program codes are adapted to be loaded and run by the processor to perform the environmental information prediction method for an autonomous vehicle as described in any one of Schemes 1 to 14 or the control method for an autonomous vehicle as described in any one of Schemes 15 to 17.

[0144] Scheme 19. A computer-readable storage medium storing a plurality of program codes, characterized in that the program codes are adapted to be loaded and run by a processor to perform the environmental information prediction method for an autonomous vehicle as described in any one of Schemes 1 to 14 or the control method for an autonomous vehicle as described in any one of Schemes 15 to 17.

[0145] Option 20. A vehicle, characterized in that the vehicle includes the control device described in Option 18. Attached Figure Description

[0146] The disclosure of this invention will become more readily understood with reference to the accompanying drawings. It will be readily understood by those skilled in the art that these drawings are for illustrative purposes only and are not intended to limit the scope of protection of this invention. Furthermore, similar numbers in the drawings are used to denote similar components, wherein:

[0147] Figure 1 This is a schematic flowchart of the main steps of an environmental information prediction method for an autonomous vehicle according to an embodiment of the present invention.

[0148] Figure 2 This is a schematic flowchart of the main steps of a control method for an autonomous vehicle according to an embodiment of the present invention;

[0149] Figure 3 This is a schematic flowchart of the main steps of a control method for an autonomous vehicle according to one embodiment of the present invention.

[0150] Figure 4 This is a schematic diagram of the encoding result of an environmental target according to one embodiment of the present invention. Detailed Implementation

[0151] Some embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0152] In the description of this invention, "module" and "processor" can include hardware, software, or a combination of both. A module can include hardware circuitry, various suitable sensors, communication ports, memory, and may also include software components, such as program code, or a combination of software and hardware. A processor can be a central processing unit, microprocessor, image processor, digital signal processor, or any other suitable processor. The processor has data and / or signal processing capabilities. The processor can be implemented in software, in hardware, or a combination of both. Non-transitory computer-readable storage media includes any suitable medium capable of storing program code, such as magnetic disks, hard disks, optical disks, flash memory, read-only memory, random access memory, etc. The term "A and / or B" means all possible combinations of A and B, such as only A, only B, or A and B. The terms "at least one A or B" or "at least one of A and B" have a similar meaning to "A and / or B" and can include only A, only B, or A and B. The singular terms "a" or "this" can also include plural forms.

[0153] See appendix Figure 1 , Figure 1 This is a schematic flowchart illustrating the main steps of an environmental information prediction method for an autonomous vehicle according to an embodiment of the present invention. Figure 1 As shown, the environmental information prediction method for autonomous vehicles in this embodiment of the invention mainly includes the following steps S101-S103.

[0154] Step S101: Construct a state transition model of the Kalman filter equation based on the vehicle motion information of the autonomous vehicle.

[0155] In this embodiment, a state transition model of the Kalman filter equation can be constructed based on the vehicle's motion information. This state transition model describes the propagation relationship between the predicted value at the previous time step and the predicted value at the current time step.

[0156] In one implementation, the vehicle motion information may include the current speed of the autonomous vehicle.

[0157] Step S102: Based on the state transition model, obtain the predicted value of the positional relationship of environmental targets in the surrounding environment of the autonomous vehicle at the current moment; wherein, the positional relationship is the relationship between the environmental targets and the position of the autonomous vehicle.

[0158] In this embodiment, the predicted positional relationships of environmental targets in the vicinity of the autonomous vehicle at the current moment can be obtained based on the state transition model. That is, the predicted value at the current moment is inferred based on the predicted value at the previous moment.

[0159] In one implementation, environmental targets may include the autonomous vehicle's own predicted trajectory points, road attributes, traffic participants, lane paths, etc. The surrounding environment may include the environment in front of and behind the autonomous vehicle. Positional relationships may include the longitudinal distance and heading of the environmental targets relative to the autonomous vehicle; where the heading is the angle between the direction of the environmental target and the autonomous vehicle; and the longitudinal distance is the distance of the environmental target relative to the vehicle along the y-axis.

[0160] Step S103: Based on the perception measurement results and predicted values ​​of the positional relationship of environmental targets at the current moment obtained by the autonomous vehicle, the predicted values ​​are updated based on the Kalman filter equation to obtain the final predicted result of the positional relationship of environmental targets at the current moment; wherein, the perception measurement results are obtained through the autonomous vehicle's motion system and / or vehicle-side perception devices and / or non-vehicle-side perception devices.

[0161] In this embodiment, the predicted values ​​can be updated based on the perception and measurement information acquired by the autonomous vehicle's motion system, vehicle-side perception devices, and non-vehicle-side perception devices, thereby obtaining the predicted results of the positional relationships of environmental targets. The update process is implemented based on the Kalman filter equation. The vehicle motion system can provide motion information of the autonomous vehicle, such as vehicle speed, yaw angle, and steering wheel angle. Vehicle-side perception devices are devices installed on the vehicle to perceive the surrounding environment, such as cameras, millimeter-wave radar, and lidar. Non-vehicle-side perception devices are devices that rely on external information to acquire perception and measurement results, such as high-precision maps, navigation maps, map matching systems, and V2X (vehicle-to-everything).

[0162] In one implementation, a predictive road model of the surrounding environment of an autonomous vehicle can be constructed according to the following formula (1):

[0163]

[0164] Where y0 is the lateral position parameter; η is the orientation angle parameter; c0 is the curvature parameter; and c1 is the rate of change of curvature parameter.

[0165] According to the above formula (1), the road tangent at a distance x and the heading from the vehicle can be obtained as shown in formula (2):

[0166]

[0167] The transfer function of the Kalman filter equation can be constructed based on the above polynomials, as shown in equations (3) and (4):

[0168] thetak =F×theta k-1 +Q (3)

[0169] z k =H×theta k +R (4)

[0170] Among them, theta k Let z be the predicted value at time k; F is the state transition matrix; Q is the system process noise matrix; z k Let be the sensing measurement result at time k; H is the transfer matrix between the sensing measurement result and the predicted value; R is the measurement noise.

[0171] theta can be expressed by the following formula (5):

[0172]

[0173] The state transition matrix can be represented by the following formula (6):

[0174]

[0175] If we assume the current vehicle speed is v and the system's operation period is T, then the state transition matrix can be represented by the following formula (7):

[0176]

[0177] The transfer matrix can be represented by the following formula (8):

[0178]

[0179] The prediction equation of the Kalman filter based on the above process can be expressed by the following formulas (9) and (10):

[0180] theta k(-) =F×theta k-1 (9)

[0181] P k(-) =F×P k-1 ×F T +Q (10)

[0182] Among them, theta k(-) To obtain the predicted value at time k based on the final predicted value based on k-1; theta k-1 P is the final predicted value for k-1; k(-) Let P be the prediction covariance matrix at time k. k-1 F is the final covariance matrix at time k-1; T This is the transpose of the state transition matrix.

[0183] The system update equations for the Kalman filter can be expressed by the following formulas (11) to (13):

[0184] K k =P k(-) ×H T / (H×P k(-) ×H T +R) (11)

[0185] theta k =theta k(-) +K k ×(z k -H×theta k(-) (12)

[0186] P k =(IK k ×H)×P k(-) ×(IK k ×H) T +K k ×R×K k T (13)

[0187] Among them, K k The Kalman gain at time k, theta k Let I be the final predicted value at time k, and let I be the identity matrix.

[0188] In one implementation, confidence levels of multiple sensing measurement results can be obtained, and the predicted values ​​can be selectively updated based on the confidence levels to obtain the final prediction result.

[0189] In one implementation, a confidence threshold can be set. If the confidence of the sensing measurement result is less than the confidence threshold, the sensing measurement result will not be used to update the predicted value. If the confidence of the sensing measurement result is greater than or equal to the confidence threshold, the sensing measurement result can be used to update the predicted value.

[0190] In one implementation, weights can be set for different levels of confidence. For example, a higher confidence level corresponds to a higher weight, while a lower confidence level corresponds to a lower weight. The predicted value can then be updated based on the perceived measurement result and its corresponding weight.

[0191] Based on steps S101-S103 above, this embodiment of the invention constructs a state transition model of the Kalman filter equation based on the autonomous vehicle's motion information. The predicted positional relationships of environmental targets in the current environment surrounding the autonomous vehicle are obtained using the state transition model. Based on the predicted values ​​and the perception measurement results of the current positional relationships, the Kalman filter equation is applied to update the predicted values, obtaining the final predicted result of the positional relationships of the environmental targets at the current moment. The perception measurement results can be obtained through the vehicle motion system, vehicle-side perception devices, and non-vehicle-side perception devices. Through this configuration, this embodiment of the invention predicts the positional relationships between environmental targets in the surrounding environment and the autonomous vehicle based on the Kalman filter model, and updates the predicted values ​​of the positional relationships of the environmental targets based on the multi-dimensional perception measurement results of the autonomous vehicle. This enables a more comprehensive prediction of the motion trends of environmental targets in the surrounding environment, exhibiting higher robustness. Simultaneously, it allows control of the autonomous vehicle based on the final predicted results of the environmental information, providing a more comprehensive, stable, and reliable control input for the autonomous vehicle's control process, achieving more effective control.

[0192] The steps for obtaining the sensing measurement results in step S103 are described in detail below.

[0193] In one embodiment of the present invention, the environmental target may include the predicted trajectory point of the autonomous vehicle within a preset aiming time, and the positional relationship includes the longitudinal distance and orientation angle of the predicted trajectory point relative to the autonomous vehicle; the perception measurement results of the predicted trajectory point can be obtained through the following steps S201 to S204.

[0194] Step S201: Obtain the vehicle speed and yaw angle of the autonomous vehicle based on the autonomous vehicle's motion system.

[0195] Step S202: Obtain the autonomous vehicle's driving radius based on the vehicle speed and yaw angle.

[0196] In this embodiment, step S202 can be further configured as follows:

[0197] Based on the assumption that autonomous vehicles are moving in a fixed circle, the vehicle's radius at each moment is obtained according to the vehicle speed and yaw angle.

[0198] In one implementation, the vehicle's driving radius can be obtained according to the following formula (14):

[0199]

[0200] Where ROC is the driving radius, host_speed is the vehicle speed, and host_yawrate is the vehicle's yaw angle.

[0201] Step S203: Obtain the vehicle trajectory of the autonomous vehicle based on the vehicle's driving radius.

[0202] Step S204: Based on the vehicle's driving trajectory, obtain the longitudinal distance and orientation angle of the predicted trajectory points before and after the vehicle's driving trajectory within the forward and backward aiming time, and use them as the perception measurement results of the predicted trajectory points.

[0203] In this embodiment, the heading values ​​and points within a certain range in front of and behind the vehicle can be obtained by multiplying the forward and backward aiming time of the vehicle's trajectory by the current vehicle speed.

[0204] [(-v*t_prd,heading(-v*t_prd)), (v*t_prd,heading(v*t_prd))]

[0205] Where t_prd is the aiming time. The above points can be used as the perception measurement results for the predicted trajectory points.

[0206] In one embodiment of the present invention, the environmental target may include road attributes, and the positional relationship may include the longitudinal distance and orientation angle of the road attributes relative to the autonomous vehicle; the perception measurement results of the road attributes can be obtained according to the following steps S301 to S303.

[0207] Step S301: Obtain road attribute perception data based on the vehicle-mounted perception devices of the autonomous vehicle.

[0208] Step S302: Discretize the perception data based on multiple discrete points of road attributes.

[0209] Step S303: Based on multiple discrete points, obtain the longitudinal distance and orientation angle of each discrete point relative to the autonomous vehicle, as the perception measurement results of road attributes. Road attributes may include information such as lane lines, curbs, and fences.

[0210] In this embodiment, a surrounding road environment model based on road attributes can be obtained according to the following formula (15):

[0211] y = c0 + c1 × x + 2 × c2 × x 2 +3×c3×x 3 (15)

[0212] Where x is the longitudinal distance of the road attribute position relative to the vehicle in the vehicle coordinate system, then according to formula (15), the heading angle at the distance x from the vehicle can be further obtained, as shown in formula (16):

[0213] headina(x)=c1+2×c2×x+3×c3×x 2 (16)

[0214] Based on formulas (15) and (16), the sensing data can be discretized to obtain discrete points of multiple road attributes. The relationship between the longitudinal distance and the orientation angle is as follows:

[0215] [(x_1, heading(x_1)), (x_2, heading(x_2)), (x_3, heading(x_3)),…, (x_n, heading(x_n))].

[0216] The predicted values ​​can be updated based on the discretized points mentioned above.

[0217] In one embodiment of the present invention, the environmental target includes traffic participants, and the positional relationship includes the longitudinal distance and orientation angle of the traffic participants relative to the autonomous vehicle. Perception measurement results of the traffic participants can be obtained according to steps S401 to S403.

[0218] Step S401: Obtain perception information of traffic participants based on the vehicle-mounted perception devices of the autonomous vehicle.

[0219] Step S402: Based on the perception information of traffic participants, obtain the displacement of traffic participants in two frames of perception information at adjacent sampling times.

[0220] Step S403: Based on the current position and displacement of the autonomous vehicle, obtain the longitudinal distance and orientation angle of the traffic participant relative to the autonomous vehicle, as the perception measurement result of the traffic participant.

[0221] In this embodiment, traffic participants may include information such as pedestrians and surrounding vehicles. The perception information of traffic participants can be obtained from the vehicle-mounted sensing device, and the orientation angle of the traffic participants can be obtained based on the displacement of the perception information between two consecutive frames, as shown in the following formulas (17) to (19):

[0222] flow dx =x current -x prev (17)

[0223] flow dy =y current -y prev (18)

[0224]

[0225] Among them, flow dx For the displacement in the x-direction, flow dy Let x be the displacement in the y-direction. current x is the x-coordinate of the current frame. prev The x-coordinate of the previous frame, y-coordinate current The y-coordinate of the current frame, y prev Here is the y-coordinate of the previous frame, and flow_angle is the orientation angle.

[0226] Based on the above formulas (17) to (19), the longitudinal distance and heading angle of all traffic participants around the vehicle can be obtained:

[0227] [(Tgt1.long, Tgt1.heading), (Tgt2.long, Tgt2.heading), (Tgt3.long, Tgt3.heading),…(Tgtn.long, Tgtn.heading)]

[0228] Where Tgt represents the traffic participant, Tgt.long represents the longitudinal distance, and Tgt.heading represents the heading angle.

[0229] In one implementation, traffic participants may include non-lane-changing vehicles, which can be determined according to steps S501 and S502.

[0230] Step S501: Obtain the speed of the surrounding vehicles of the autonomous vehicle and the angle between the current time and the previous time of the surrounding vehicles.

[0231] Step S502: Identify surrounding vehicles whose speed is greater than the preset speed and whose angle is less than the preset angle as non-lane-changing vehicles.

[0232] In this embodiment, it can be determined whether a vehicle is not changing lanes based on the speed of surrounding vehicles and the angle between the current time and the previous time.

[0233] In one embodiment of the present invention, the environmental target may include the lane path, and the positional relationship may include the longitudinal distance and orientation angle between the lane path and the autonomous vehicle; the perception measurement results of the lane path can be obtained according to the following steps S601 to S603.

[0234] Step S601: Obtain the driving path of the autonomous vehicle based on the non-vehicle-side perception devices of the autonomous vehicle.

[0235] Step S602: Based on the current position and driving path of the autonomous vehicle, obtain the lane path discrete points of the lane where the autonomous vehicle is currently located and the surrounding lanes.

[0236] Step S603: Use the longitudinal distance and orientation angle of the discrete points of the lane path relative to the autonomous vehicle as the perception measurement results of the lane path.

[0237] In this embodiment, the driving path of the autonomous vehicle can be obtained using non-vehicle-side sensing devices, which may include navigation maps and high-precision maps. When the vehicle is traveling within the coverage area of ​​the high-precision map, high-precision map information of the driving path can be obtained by fusing the navigation map and the high-precision map. This information includes detailed information on path level, road level, and lane level. Based on the current position of the autonomous vehicle, lane level information in front of and behind the vehicle can be obtained, such as the current lane ID, the longitudinal distance and orientation angle of the current and surrounding lanes relative to the vehicle, specifically:

[0238] [(pos1.long, pos1.heading), (pos 2.long, pos2.heading), (pos 3.long, pos3.heading),…, (pos n.long, pos n.heading)]

[0239] Where pos.long is the longitudinal distance between discrete points on the lane path, and pos.heading is the orientation angle of the discrete points on the lane path.

[0240] In one implementation, when roads merge, the discrete points of the lane path behind the autonomous vehicle can be obtained based on the current location; and the longitudinal distance and orientation angle of the discrete points of the lane path behind the autonomous vehicle relative to the autonomous vehicle can be used as the perception measurement results of the lane path.

[0241] In one embodiment of the present invention, before step S102, the present invention may further include steps S104 and S105.

[0242] Step S104: Initialize the perception measurement information of environmental targets.

[0243] Step S105: Based on the initialization results, determine the effectiveness of the sensing measurement information acquisition device.

[0244] In this embodiment, before predicting the positional relationship of environmental targets, the sensing measurement information of the environmental targets can be initialized first, and the validity of the sensing measurement acquisition device can be determined based on the initialization result. If the determination is successful, the step of predicting the positional relationship of the environmental targets can then be performed.

[0245] In one implementation, when the acquisition device is a vehicle-mounted perception device of an autonomous vehicle, the overall confidence level of the vehicle-mounted perception device can be obtained based on the initial confidence level of the vehicle-mounted perception device and environmental information; and the effectiveness of the vehicle-mounted perception device can be determined based on the overall confidence level.

[0246] In this embodiment, if the device acquiring the perception measurement information is a vehicle-mounted perception device, such as a camera, LiDAR, or millimeter-wave radar, the initial confidence level of the vehicle-mounted perception device can be used to determine whether it can predict the positional relationships of environmental targets. Then, the overall confidence level of the perception device is obtained based on the environmental information, and the effectiveness of the vehicle-mounted perception device is determined based on this overall confidence level. The initial confidence level refers to the default confidence level of the lane perception device. The overall confidence level is the confidence level obtained by combining the initial confidence level and the environmental information. Taking lane lines as an example, the environmental information may include information such as their length, clarity, curvature, and whether they intersect.

[0247] In one implementation, when the acquisition device is a map matching system, the validity is determined based on the communication between different maps in the map matching system, whether the positioning system of the autonomous vehicle is working properly, and whether the current road is covered by a high-precision map.

[0248] In one embodiment of the present invention, step S103 may further include steps S1031 and S1032:

[0249] Step S1031: Update the predicted value based on the perception measurement results of the moving target, and obtain the updated predicted value of the moving target.

[0250] Step S1032: Obtain the final prediction result based on the updated predicted value of the moving target and the positional relationship between the stationary target and the autonomous vehicle.

[0251] In this embodiment, during the prediction and updating of the positional relationships of environmental targets, stationary targets are not included in the discussion; only moving targets are predicted and updated. Then, the final prediction result is obtained based on the updated predicted value of the moving targets and the positional relationship between the stationary targets and the self-workshop.

[0252] In one implementation, when the traffic participant is a non-lane-changing vehicle, step S103 may further include the following steps S1033 to S1035:

[0253] Step S1033: Obtain the mean and variance of the orientation angles based on the multiple orientation angles of non-lane-changing vehicles.

[0254] In this embodiment, the mean and variance of the orientation angle can be obtained based on multiple orientation angles of non-lane-changing vehicles.

[0255] Step S1034: Determine the valid non-lane-changing vehicles based on the mean and the preset mean.

[0256] In this embodiment, a preset average value can be set, and the average orientation angle value can be compared with the preset average value to determine the valid non-lane-changing vehicles.

[0257] Step S1035: Update the predicted values ​​based on the longitudinal distance and orientation angle of the valid non-lane-changing vehicles.

[0258] In this embodiment, step S1035 may further include steps S10351 and S10352:

[0259] Step S10351: Use the variance of the orientation angle of non-lane-changing vehicles as noise for updating the Kalman filter equation.

[0260] Step S10352: Update the predicted values ​​based on the noise, as well as the longitudinal distance and orientation angle of the effective non-lane-changing vehicles.

[0261] In this embodiment, the variance of the orientation angle of non-lane-changing vehicles can be used as noise to update the Kalman filter equation, and the predicted values ​​of traffic participants can be updated based on the longitudinal distance and orientation angle of effective non-lane-changing vehicles.

[0262] Furthermore, the present invention also provides a control method for an autonomous vehicle.

[0263] See appendix Figure 2 , Figure 2 This is a schematic flowchart illustrating the main steps of a control method for an autonomous vehicle according to an embodiment of the present invention. Figure 2 As shown, the control method for autonomous vehicles in this embodiment of the invention mainly includes the following steps S701-S702.

[0264] Step S701: According to the environmental information prediction method for autonomous vehicles in the above embodiment of the autonomous vehicle environmental information prediction method, obtain the final prediction result of environmental targets in the surrounding environment of the autonomous vehicle.

[0265] In this embodiment, the final prediction result of environmental targets in the surrounding environment can be obtained through the environmental information prediction method of autonomous vehicles.

[0266] Step S702: Control the autonomous vehicle based on the final prediction result.

[0267] In this embodiment, the behavior planning of autonomous vehicles can be realized based on the final prediction results of environmental targets, so as to further realize the control process of autonomous vehicles, such as steering control, drive control, braking control, etc.

[0268] In one embodiment, step S702 may further include steps S7021 and S7022.

[0269] Step S7021: Encode the environmental targets based on the final prediction results.

[0270] Step S7022: Control the autonomous vehicle according to the coding results.

[0271] In this embodiment, please refer to the appendix. Figure 4 , Figure 4 This is a schematic diagram illustrating the encoding result of an environmental target according to one embodiment of the present invention. Figure 4 As shown, environmental targets can be encoded based on different environmental targets and the final prediction results of the environmental targets, thereby enabling the control of autonomous vehicles based on the encoding results.

[0272] In one implementation, the environmental target is the vehicles surrounding the autonomous vehicle, and step S7021 may further include steps S70211 and S70212.

[0273] Step S70211: Obtain the orientation angles of the surrounding vehicles based on the final prediction results of the surrounding vehicles.

[0274] Step S70212: Encode surrounding vehicles based on the orientation angle and a preset included angle threshold.

[0275] In this embodiment, it can be determined whether the orientation angle of the surrounding vehicles is less than the included angle threshold; when it is less than the included angle threshold, the surrounding vehicles can be encoded as non-lane-changing attributes; when the included angle is greater than or equal to the included angle threshold, the surrounding vehicles can be encoded as lane-changing attributes.

[0276] In one implementation, see Appendix Figure 3 , Figure 3 This is a schematic flowchart illustrating the main steps of a control method for an autonomous vehicle according to one embodiment of the present invention. Figure 3 As shown, the control method for an autonomous vehicle may include the following steps S801 to S823.

[0277] Step S801: Initialize the vehicle motion system.

[0278] In this embodiment, the vehicle motion system can be initialized first.

[0279] Step S802: Is the data from the vehicle motion system valid? If yes, proceed to step S803; if no, proceed to step S801.

[0280] In this embodiment, it can be determined whether the data of the vehicle motion system (such as vehicle speed, yaw angle, etc.) is valid. If it is valid, step S803 can be performed, and the valid data of the vehicle motion system can also be applied to step S822 to update the predicted value.

[0281] Step S803: Is the vehicle traveling along the lane line? If yes, proceed to step S804; if no, proceed to step S801.

[0282] In this embodiment, it can be determined whether the vehicle is traveling along the lane line.

[0283] Step S804: Update the environment model before and after based on the fixed circle property (fixed circle motion assumption).

[0284] In this embodiment, step S804 is similar to the method described in steps S202 and S203 above, and will not be repeated here for the sake of simplicity.

[0285] Step S805: Obtain longitudinal distance and heading information (orientation angle) based on the pre-aiming time, and proceed to step S822.

[0286] In this embodiment, step S805 is similar to the method described in step S204 above, and will not be repeated here for the sake of simplicity.

[0287] Step S806: Initialize the vehicle perception system (vehicle-side perception device).

[0288] In this embodiment, the vehicle perception system can be initialized first.

[0289] Step S807: Check if the vehicle perception system diagnosis is successful; if yes, proceed to steps S808 and S812; if no, proceed to step S806.

[0290] In this embodiment, it can be determined that the diagnostic of the vehicle perception system is successful.

[0291] Step S808: Detect whether the lane lines, curbs, and fences are qualified; if yes, proceed to step S809; if no, proceed to step S807.

[0292] In this embodiment, it is possible to verify whether lane lines, curbs, fences, etc., have passed inspection.

[0293] Step S809: Are there any lane intersections, mergings, or loss? If not, proceed to step S810; if yes, proceed to step S807.

[0294] In this embodiment, it is possible to determine whether there are situations of lane crossing, merging, or loss.

[0295] Step S810: Discretize the sensing data.

[0296] In this embodiment, step S810 is similar to the method described in step S302 above, and will not be repeated here for the sake of simplicity.

[0297] Step S811: Select the valid lane lines, curbs, fence information (longitudinal distance) and heading information (orientation angle), and proceed to step S822.

[0298] In this embodiment, step S811 is similar to the method described in step S303 above, and will not be repeated here for the sake of simplicity.

[0299] Step S812: Sensing the movement information of traffic participants.

[0300] In this embodiment, step S812 is similar to the method described in step S401 above, and will not be repeated here for the sake of simplicity.

[0301] Step S813: Determine whether the traffic participant is a moving target. If yes, proceed to step S814; otherwise, proceed to step S812.

[0302] In this embodiment, it can be determined whether the traffic participant is a moving target. If it is a moving target, step S814 is executed. Valid moving targets are also used for updating the predicted value in step S822.

[0303] Step S814: Update the obtained motion heading information and obtain the mean and variance.

[0304] In this embodiment, step S814 is similar to the method described in steps S402 and S403 above, and will not be repeated here for the sake of simplicity.

[0305] Step S815: Select valid moving target orientation (longitudinal distance) and heading information, then proceed to step S822.

[0306] In this embodiment, a valid moving target can be selected for prediction.

[0307] Step S816: Initialize the non-vehicle sensing system (device).

[0308] In this embodiment, the non-vehicle sensing system can be initialized.

[0309] Step S817: Has the system diagnosis passed? If yes, proceed to step S818; if no, proceed to step S816.

[0310] In this embodiment, it can be determined whether the system diagnosis has passed.

[0311] Step S818: Match vehicle location with high-precision map information.

[0312] Step S819: Match the navigation map with future route information.

[0313] Step S820: Output the map information of the road segments before and after the current vehicle.

[0314] In this embodiment, the methods described in steps S818, S819 and S820 are similar to those in step S601 described above, and will not be repeated here for the sake of simplicity.

[0315] Step S821: Extract the discrete points of the lane path from the high-precision map.

[0316] In this embodiment, the method described in step S821 is similar to that in step S602 described above, and will not be repeated here for the sake of describing the detection.

[0317] Step S822: Update the predicted values ​​of environmental targets according to the Kalman state equation to obtain the type and confidence level of the environmental targets.

[0318] In this embodiment, the predicted values ​​of environmental targets can be updated according to the Kalman equation to obtain the type and confidence level of the environmental targets.

[0319] Step S823: Based on the relationship between the vehicle and the environmental target, obtain the coding result of the environmental target.

[0320] In this embodiment, the method described in step S823 is similar to that in step S7021 described above, and will not be repeated here for the sake of simplicity.

[0321] It should be noted that although the steps in the above embodiments are described in a specific order, those skilled in the art will understand that, in order to achieve the effects of the present invention, different steps do not necessarily have to be executed in such an order. They can be executed simultaneously (in parallel) or in other orders, and these variations are all within the scope of protection of the present invention.

[0322] It should be noted that the data involved in the embodiments of this disclosure (including but not limited to data used for analysis, stored data, displayed data, vehicle usage data, vehicle-collected data, etc.) are all data that has been fully authorized by all parties. The actions of acquiring and collecting data involved in the embodiments of this disclosure are all performed after authorization by the user, the object, or after full authorization by all parties.

[0323] Those skilled in the art will understand that all or part of the processes in the method of the above embodiment of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or some intermediate form. The computer-readable storage medium can include any entity or device capable of carrying the computer program code, a medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory, a random access memory, an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc. It should be noted that the content included in the computer-readable storage medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable storage medium does not include electrical carrier signals and telecommunication signals.

[0324] Furthermore, the present invention also provides a control device. In one embodiment of the control device according to the present invention, the control device includes a processor and a storage device. The storage device can be configured to store a program for executing the environmental information prediction method for an autonomous vehicle according to the above-described method embodiments. The processor can be configured to execute the program in the storage device, which includes, but is not limited to, a program for executing the environmental information prediction method for an autonomous vehicle according to the above-described method embodiments. For ease of explanation, only the parts related to the embodiments of the present invention are shown; for specific technical details not disclosed, please refer to the method section of the embodiments of the present invention. The control device can be a control device device comprising various electronic devices.

[0325] In embodiments of the present invention, the control device may be a control device device comprising various electronic devices. In some possible implementations, the control device may include multiple storage devices and multiple processors. The program executing the environmental information prediction method for autonomous vehicles in the above-described method embodiments can be divided into multiple subroutines, each subroutine can be loaded and run by a processor to execute different steps of the environmental information prediction method for autonomous vehicles in the above-described method embodiments. Specifically, each subroutine can be stored in different storage devices, and each processor can be configured to execute programs in one or more storage devices to jointly implement the environmental information prediction method for autonomous vehicles in the above-described method embodiments, that is, each processor executes different steps of the environmental information prediction method for autonomous vehicles in the above-described method embodiments to jointly implement the environmental information prediction method for autonomous vehicles in the above-described method embodiments.

[0326] The aforementioned multiple processors can be processors deployed on the same device. For example, the aforementioned control device can be a high-performance device composed of multiple processors, and the aforementioned multiple processors can be processors configured on that high-performance device. Alternatively, the aforementioned multiple processors can also be processors deployed on different devices. For example, the aforementioned control device can be a server cluster, and the aforementioned multiple processors can be processors on different servers within the server cluster.

[0327] Furthermore, the present invention also provides a computer-readable storage medium. In one embodiment of the computer-readable storage medium according to the present invention, the computer-readable storage medium can be configured to store a program for executing the environmental information prediction method for an autonomous vehicle of the above-described method embodiments. This program can be loaded and run by a processor to implement the environmental information prediction method for the autonomous vehicle of the above-described method. For ease of explanation, only the parts related to the embodiments of the present invention are shown; for specific technical details not disclosed, please refer to the method section of the embodiments of the present invention. The computer-readable storage medium can be a storage device comprising various electronic devices. Optionally, in the embodiments of the present invention, the computer-readable storage medium is a non-transitory computer-readable storage medium.

[0328] Furthermore, the present invention also provides a control device. In one embodiment of the control device according to the present invention, the control device includes a processor and a storage device. The storage device can be configured to store a program for executing the control method of the autonomous vehicle of the above-described method embodiments. The processor can be configured to execute the program in the storage device, which includes, but is not limited to, a program for executing the control method of the autonomous vehicle of the above-described method embodiments. For ease of explanation, only the parts related to the embodiments of the present invention are shown; for specific technical details not disclosed, please refer to the method section of the embodiments of the present invention. The control device can be a control device device comprising various electronic devices.

[0329] In embodiments of the present invention, the control device may be a control device device comprising various electronic devices. In some possible implementations, the control device may include multiple storage devices and multiple processors. The program for executing the control method of the autonomous vehicle in the above-described method embodiments can be divided into multiple subroutines, each subroutine can be loaded and run by a processor to execute different steps of the control method of the autonomous vehicle in the above-described method embodiments. Specifically, each subroutine can be stored in different storage devices, and each processor can be configured to execute programs in one or more storage devices to jointly implement the control method of the autonomous vehicle in the above-described method embodiments, that is, each processor executes different steps of the control method of the autonomous vehicle in the above-described method embodiments to jointly implement the control method of the autonomous vehicle in the above-described method embodiments.

[0330] The aforementioned multiple processors can be processors deployed on the same device. For example, the aforementioned control device can be a high-performance device composed of multiple processors, and the aforementioned multiple processors can be processors configured on that high-performance device. Alternatively, the aforementioned multiple processors can also be processors deployed on different devices. For example, the aforementioned control device can be a server cluster, and the aforementioned multiple processors can be processors on different servers within the server cluster.

[0331] Furthermore, the present invention also provides a computer-readable storage medium. In one embodiment of the computer-readable storage medium according to the present invention, the computer-readable storage medium can be configured to store a program for executing the control method of the autonomous vehicle of the above-described method embodiments. This program can be loaded and run by a processor to implement the control method of the autonomous vehicle described above. For ease of explanation, only the parts related to the embodiments of the present invention are shown; for specific technical details not disclosed, please refer to the method section of the embodiments of the present invention. The computer-readable storage medium can be a storage device comprising various electronic devices. Optionally, in the embodiments of the present invention, the computer-readable storage medium is a non-transitory computer-readable storage medium.

[0332] Furthermore, the present invention also provides a vehicle. In one embodiment of the vehicle according to the present invention, the vehicle may include a control device as described in the control device embodiment.

[0333] Furthermore, it should be understood that since the various modules are only provided to illustrate the functional units of the device of the present invention, the physical devices corresponding to these modules may be the processor itself, or a part of the processor's software, hardware, or a combination of software and hardware. Therefore, the number of modules shown in the figures is merely illustrative.

[0334] Those skilled in the art will understand that the various modules in the device can be adaptively split or combined. Such splitting or combining of specific modules will not cause the technical solution to deviate from the principles of the present invention; therefore, the technical solutions after splitting or combining will fall within the protection scope of the present invention.

[0335] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after such changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A method for predicting environmental information for autonomous vehicles, characterized in that, The method includes: Based on the vehicle motion information of the autonomous vehicle, a state transition model of the Kalman filter equation is constructed; Based on the state transition model, the predicted values ​​of the positional relationships of environmental targets in the surrounding environment of the autonomous vehicle at the current moment are obtained; wherein, the positional relationship is the relationship between the environmental targets and the position of the autonomous vehicle. Based on the perception measurement results of the positional relationship of the environmental targets at the current moment obtained by the autonomous vehicle and the predicted value, the predicted value is updated based on the Kalman filter equation to obtain the final predicted result of the positional relationship of the environmental targets at the current moment. The perception measurement results are obtained through the autonomous vehicle's motion system and / or vehicle-side perception devices and / or non-vehicle-side perception devices. The environmental targets include at least one of the following within a preset pre-aiming time: the predicted trajectory points of the autonomous vehicle, road attributes, traffic participants, and lane paths.

2. The environmental information prediction method for autonomous vehicles according to claim 1, characterized in that, The positional relationship includes the longitudinal distance and orientation angle of the predicted trajectory point relative to the autonomous vehicle; The method further includes: Based on the autonomous vehicle's motion system, the vehicle speed and yaw angle of the autonomous vehicle are obtained; The autonomous vehicle's driving radius is obtained based on the vehicle speed and the yaw angle. Based on the autonomous vehicle's driving radius, the vehicle's driving trajectory is obtained; Based on the vehicle's driving trajectory, the longitudinal distance and orientation angle of the predicted trajectory points before and after the vehicle's driving trajectory are obtained within the preceding and following aiming time, and are used as the perception measurement results of the predicted trajectory points.

3. The environmental information prediction method for autonomous vehicles according to claim 2, characterized in that, The step of "obtaining the autonomous vehicle's driving radius based on the vehicle speed and yaw angle" includes: Based on the assumption that the autonomous vehicle is moving in a fixed circle, the vehicle's driving radius at each moment is obtained according to the vehicle speed and yaw angle.

4. The environmental information prediction method for autonomous vehicles according to claim 1, characterized in that, The positional relationship includes the longitudinal distance and orientation angle of the road attributes relative to the autonomous vehicle; The method further includes: Based on the vehicle-mounted perception devices of the autonomous vehicle, the perception data of the road attributes are obtained; The perceived data is discretized based on multiple discrete points of the road attributes; Based on the plurality of discrete points, the longitudinal distance and orientation angle of the discrete points relative to the autonomous vehicle are obtained as the perception measurement results of the road attributes.

5. The environmental information prediction method for autonomous vehicles according to claim 1, characterized in that, The positional relationship includes the longitudinal distance and orientation angle of the traffic participant relative to the autonomous vehicle; The method further includes: The perception information of the traffic participants is obtained based on the vehicle-mounted perception device of the autonomous vehicle. Based on the perception information of traffic participants, the displacement of the traffic participants in two frames of perception information at adjacent sampling times is obtained; Based on the current position and displacement of the autonomous vehicle, the longitudinal distance and orientation angle of the traffic participant relative to the autonomous vehicle are obtained as the perception measurement results of the traffic participant.

6. The environmental information prediction method for autonomous vehicles according to claim 5, characterized in that, The traffic participants include vehicles that are not changing lanes; The method further includes: The speeds of surrounding vehicles and the angles between the current and previous times of the surrounding vehicles are obtained for the autonomous vehicle. The surrounding vehicles whose speed is greater than a preset speed and whose angle is less than a preset angle are considered as the non-lane-changing vehicles.

7. The environmental information prediction method for autonomous vehicles according to claim 6, characterized in that, The step of "updating the predicted value based on the Kalman filter equation according to the perception measurement results of the autonomous vehicle at the current moment and the predicted value" includes: Based on the multiple orientation angles of the non-lane-changing vehicles, obtain the mean and variance of the orientation angles; Based on the mean and the preset mean, determine the valid non-lane-changing vehicles; The predicted value is updated based on the longitudinal distance and orientation angle of the effective non-lane-changing vehicles.

8. The environmental information prediction method for autonomous vehicles according to claim 7, characterized in that, The step of "updating the predicted value based on the longitudinal distance and orientation angle of the effective non-lane-changing vehicles" includes: The variance of the orientation angle of the non-lane-changing vehicles is used as noise to update the Kalman filter equation; The predicted values ​​are updated based on the noise, as well as the longitudinal distance and orientation angle of the effective non-lane-changing vehicles.

9. The environmental information prediction method for autonomous vehicles according to claim 1, characterized in that, The positional relationship includes the longitudinal distance and orientation angle between the lane path and the autonomous vehicle; The method further includes: The driving path of the autonomous vehicle is obtained based on the non-vehicle-side perception devices of the autonomous vehicle. Based on the current location of the autonomous vehicle and the driving path, obtain the lane path discrete points of the lane where the autonomous vehicle is currently located and the surrounding lanes. The longitudinal distance and orientation angle of the discrete points of the lane path relative to the autonomous vehicle are used as the perception measurement results of the lane path.

10. The environmental information prediction method for autonomous vehicles according to claim 9, characterized in that, The step of "obtaining the lane path discrete points of the lane where the autonomous vehicle is currently located and the surrounding lanes based on the current position of the autonomous vehicle and the driving path" includes: When roads merge, the discrete points of the lane path behind the autonomous vehicle are obtained based on the current location; The step of "using the longitudinal distance and orientation angle of the discrete points of the lane path relative to the autonomous vehicle as the perception measurement result of the lane path" includes: The longitudinal distance and orientation angle of the discrete points of the rear lane path relative to the autonomous vehicle are used as the perception measurement results of the lane path.

11. The environmental information prediction method for autonomous vehicles according to claim 1, characterized in that, Before the step of "obtaining the predicted values ​​of the positional relationships of environmental targets in the surrounding environment of the autonomous vehicle at the current moment based on the state transition model", the method further includes: Initialize the perceived measurement information of the environmental target; Based on the initialization results, the effectiveness of the sensing measurement information acquisition device is determined.

12. The environmental information prediction method for autonomous vehicles according to claim 11, characterized in that, The step of "determining the validity of the device for acquiring the sensing measurement information" includes: When the acquisition device is the vehicle-side perception device of the autonomous vehicle, the comprehensive confidence level of the vehicle-side perception device is obtained based on the initial confidence level of the vehicle-side perception device and the environmental information. Based on the comprehensive confidence level, the effectiveness of the vehicle-mounted sensing device is determined; and / or, When the acquisition device is a map matching system, the validity is determined based on the communication between different maps in the map matching system, whether the positioning system of the autonomous vehicle is working properly, and whether the current road is covered by a high-precision map.

13. The environmental information prediction method for autonomous vehicles according to claim 1, characterized in that, Before the step of "updating the predicted values ​​based on the perception measurement results of the positional relationships of the environmental targets at the current moment obtained by the autonomous vehicle and the predicted values, and obtaining the final predicted results of the positional relationships of the environmental targets at the current moment", the method includes: Obtain the confidence level of the perception measurement results; Based on the confidence level, the predicted value is selectively updated using the perceived measurement results to obtain the final prediction result.

14. The environmental information prediction method for autonomous vehicles according to claim 1, characterized in that, The environmental targets include moving targets and stationary targets. The step of "updating the predicted values ​​based on the perception measurement results of the positional relationships of the environmental targets at the current moment obtained by the autonomous vehicle and the predicted values, and obtaining the final predicted results of the positional relationships of the environmental targets at the current moment" includes: The predicted value is updated based on the perception measurement results of the moving target to obtain the updated predicted value of the moving target; The final prediction result is obtained based on the updated predicted value of the moving target and the positional relationship between the stationary target and the autonomous vehicle.

15. A control method for an autonomous vehicle, characterized in that, The method includes: According to any one of claims 1 to 14, the method for predicting environmental information of an autonomous vehicle obtains the final prediction result of environmental targets in the surrounding environment of the autonomous vehicle. The autonomous vehicle is controlled based on the final prediction result.

16. The control method for an autonomous vehicle according to claim 15, characterized in that, The step of "controlling the autonomous vehicle based on the final prediction result" includes: The environmental target is encoded based on the final prediction result; The autonomous vehicle is controlled based on the encoding result.

17. The control method for an autonomous vehicle according to claim 16, characterized in that, The environmental target is the vehicles surrounding the autonomous vehicle; the step of "encoding the environmental target based on the final prediction result" includes: Based on the final prediction results of the surrounding vehicles, obtain the orientation angle of the surrounding vehicles; The surrounding vehicles are encoded based on the orientation angle and a preset included angle threshold.

18. A control device comprising at least one processor and at least one storage device, said storage device being adapted to store a plurality of program codes, characterized in that, The program code is adapted to be loaded and run by the processor to perform the environmental information prediction method for an autonomous vehicle according to any one of claims 1 to 14 or the control method for an autonomous vehicle according to any one of claims 15 to 17.

19. A computer-readable storage medium storing a plurality of program codes, characterized in that, The program code is adapted to be loaded and run by a processor to perform the environmental information prediction method for an autonomous vehicle as described in any one of claims 1 to 14 or the control method for an autonomous vehicle as described in any one of claims 15 to 17.

20. A vehicle, characterized in that, The vehicle includes the control device as described in claim 18.

Images