Automotive autonomous driving system and behavior safety method

Abstract

The application discloses an automatic driving system and a behavior safety method for a vehicle. The system comprises a main channel and a backup channel in parallel, and is composed of a first environment sensing module, a second environment sensing module, an environment fusion module, a trajectory generation module, a backup trajectory generation module, a safety verification module, a state switching module and a trajectory execution module. The first environment sensing module and the second environment sensing module serve as the environment sensing of the main channel, and the robustness of the environment sensing is improved. The backup trajectory generation module is combined with the second environment sensing module to output a backup trajectory, and the state switching module is facilitated to select. The environment fusion module serves as the environment sensing fusion of the main channel, and the environment sensing result is improved. The safety verification module verifies the main trajectory and the backup trajectory, and the output of the trajectory does not need to be corrected, and the influence of the safety verification and the trajectory correction on the availability of the system is reduced. The state switching module realizes the redundancy of the trajectory output, and the reliability of the system is improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of intelligent driving, and relates to the safety of an intelligent driving system, in particular to an automatic driving system for a vehicle and a behavior safety method. BACKGROUND

[0002] In the prior art, in order to realize safe behavior of an automatic driving vehicle, similar technical solutions are usually adopted, such as the vehicle and its safe driving method and device disclosed in CN111114540B. The specific principle is as follows: first, an environment perception module is responsible for collecting relevant information from the surrounding environment of the vehicle; second, the environment perception module outputs the collected relevant information to a trajectory generation module and a safety verification and trajectory correction module; the safety verification and trajectory correction module verifies and corrects the trajectory output by the trajectory generation module to ensure that a safe driving trajectory is generated, and outputs the safe trajectory to a trajectory execution module; finally, the trajectory execution module controls the vehicle to drive according to the safe trajectory output by the upstream module.

[0003] In the above technical solution, the execution logic of the safety verification and trajectory correction module is usually constructed based on rules, that is, based on the positional relationship between the ego vehicle and the surrounding objects, and the acceleration and deceleration capabilities of the ego vehicle and the surrounding objects, the safety constraints that the ego vehicle needs to ensure for safety are inversely deduced, which has the following defects:

[0004] 1. Influence of the environment perception module:

[0005] As an important component of the automatic driving system, the performance of the environment perception module directly affects the effect of the trajectory generation and safety verification and trajectory correction modules.

[0006] If the environment perception module has insufficient functions, such as high false detection and missed detection under specific lighting, weather conditions, or for specific objects, even after processing by the safety verification and trajectory correction module, the output trajectory may not be safe.

[0007] This is because the erroneous information of the environment perception module is passed to the downstream modules, causing the entire system to make decisions based on inaccurate data, thereby increasing the safety risk.

[0008] 2. Severity of environment perception module failure:

[0009] If the environment perception module has a serious failure, such as module crash, the downstream trajectory generation and safety verification and trajectory correction modules will not work normally.

[0010] In this case, the system cannot output a safe vehicle trajectory, and thus cannot ensure the safe driving of the vehicle.

[0011] This indicates that the safety of the entire autonomous driving system largely depends on the stability and reliability of the environmental perception module.

[0012] 3. Mismatch between preset rules and actual scenarios:

[0013] The safety verification and trajectory correction module usually works based on preset rules, which are formulated according to general scenarios and assumptions.

[0014] However, during actual driving, various complex and variable scenarios may be encountered, which may not fully match the preset rules.

[0015] If the preset rules do not match the actual scenarios, the safety verification and trajectory correction module may misjudge unsafe scenarios as safe or vice versa, affecting the overall usability and accuracy of the intelligent driving system.

[0016] In summary, although the safety verification and trajectory correction module is based on rules in its execution logic, this method has obvious shortcomings that need to be continuously optimized and improved in practical applications. To improve the safety and reliability of the autonomous driving system, the development and testing of the environmental perception module need to be strengthened to ensure its accurate and stable perception of the surrounding environment; at the same time, the rules and algorithms of the safety verification and trajectory correction module need to be continuously improved and optimized to adapt to more complex and variable actual driving scenarios. SUMMARY

[0017] To solve the above problems, the main purpose of the present application is to design an automobile automatic driving system and behavior safety method, which is composed of a parallel main channel and a backup channel through a first environmental perception module, a second environmental perception module, an environmental fusion module, a trajectory generation module, a backup trajectory generation module, a safety verification module, a state switching module, and a trajectory execution module, to solve the problem of trajectory output error or inability to output trajectory caused by functional deficiencies of the environmental perception module, and the poor accuracy of the system caused by safety verification and trajectory correction misjudgment.

[0018] To achieve the above purpose, the present application adopts the following technical solutions:

[0019] An automobile automatic driving system, which comprises a parallel main channel and a backup channel; wherein the main channel and the backup channel are composed of a first environmental perception module, a second environmental perception module, an environmental fusion module, a trajectory generation module, a backup trajectory generation module, a safety verification module, a state switching module, and a trajectory execution module.

[0020] The main channel is composed of a first environmental perception module, a second environmental perception module, an environmental fusion module, a trajectory generation module, a safety verification module, and a state switching module.

[0021] The first environment perception module and the second environment perception module respectively receive sensor data and transmit perception results to the environment fusion module, the trajectory generation module receives the output of the environment fusion module, generates a main trajectory, and outputs the main trajectory to the safety verification module and the state switching module;

[0022] The backup channel is composed of the second environment perception module, the backup trajectory generation module, the safety verification module, and the state switching module;

[0023] The second environment perception module receives sensor data and transmits perception results to the backup trajectory generation module, the backup trajectory generation module generates a backup trajectory according to the received perception results, and outputs the backup trajectory to the safety verification module and the state switching module;

[0024] The state switching module determines to adopt the main trajectory or the backup trajectory based on the verification result output by the safety verification module, and outputs the adopted trajectory;

[0025] The trajectory execution module receives and executes the trajectory output by the state switching module.

[0026] As a further description of the present application, the input data of the first environment perception module and the second environment perception module is original sensor data, including camera data, laser radar data, and millimeter wave radar data, and the output data is perception results, including the position, speed, and category of objects in the environment, and road information;

[0027] The first environment perception module processes sensor data based on a deep learning model to obtain perception results;

[0028] The second environment perception module processes sensor data based on a traditional algorithm and a lightweight deep learning model to obtain perception results.

[0029] As a further description of the present application, the environment fusion module fuses the perception results of the first environment perception module and the second environment perception module as input to obtain fused perception results, and outputs the fused perception results;

[0030] The fusion method of the environment fusion module includes one of a target level fusion algorithm, a probability fusion method, and a weighted voting mechanism.

[0031] As a further description of the present application, the trajectory generation module takes the fused perception results output by the environment fusion module and the vehicle navigation target as input, outputs a planned driving path of the vehicle, and defines the path as the main trajectory;

[0032] The trajectory generation module generates the main trajectory by using a hierarchical planning architecture, a path planning algorithm and a speed planning algorithm.

[0033] As a further description of the application, the backup trajectory generation module takes the perception result of the second environment perception module as input and outputs a planned driving path of the vehicle, which is defined as the backup trajectory.

[0034] The backup trajectory generation module generates the backup trajectory by using a pure pursuit algorithm, a safety distance setting and a preset emergency operation.

[0035] As a further description of the application, the safety verification module takes the main trajectory generated by the trajectory generation module, the backup trajectory generated by the backup trajectory generation module and the perception result as input, performs safety verification and outputs the verification result of the main trajectory and the backup trajectory.

[0036] The safety verification mode of the safety verification module is one or more of rule-based collision detection, space-time safety corridor verification and probability risk assessment.

[0037] As a further description of the application, the state switching module selects the trajectory to be finally executed according to the verification result of the safety verification module; the input of the state switching module includes the main trajectory, the backup trajectory and the safety verification result, and the output is the trajectory to be finally executed.

[0038] The selection process of the state switching module includes a normal working mode, a condition for triggering switching, switching logic and special case processing.

[0039] As a further description of the application, the trajectory execution module realizes the movement of the vehicle according to the output trajectory of the state switching module through a control instruction.

[0040] The implementation mode of the trajectory execution module adopts a lateral controller, a longitudinal controller and an execution monitoring mechanism.

[0041] An automotive automatic driving behavior safety method based on the above system, including a main trajectory generation process, a backup trajectory generation process and a main trajectory and backup trajectory control process.

[0042] The main trajectory generation process is implemented based on the main channel and includes the following steps:

[0043] S11: The first environment perception module and the second environment perception module respectively receive sensor data and obtain perception results.

[0044] S12: The first environment perception module and the second environment perception module output the perception results to the environment fusion module, fuse the perception results of the first environment perception module and the perception results of the second environment perception module, and obtain the fused perception results.

[0045] S13: The environment fusion module outputs the fused perception result to the trajectory generation module for main trajectory generation.

[0046] S14: The trajectory generation module outputs the main trajectory to the safety verification module and the state switching module for verifying the safety of the main trajectory.

[0047] The backup trajectory generation process is implemented based on a backup channel and includes the following steps:

[0048] S21: The second environment perception module receives sensor data and obtains a perception result.

[0049] S22: The second environment perception module outputs the perception result to the backup trajectory generation module for backup trajectory generation.

[0050] S23: The trajectory generation module outputs the backup trajectory to the safety verification module and the state switching module for verifying the safety of the backup trajectory.

[0051] The main trajectory and backup trajectory control process includes the following steps:

[0052] S31: The safety verification module outputs the verification result of the main trajectory and the backup trajectory, the main trajectory, and the backup trajectory to the state switching module.

[0053] S32: The state switching module selects and outputs the main trajectory or the backup trajectory to the trajectory execution module according to the verification result.

[0054] S33: The trajectory execution module receives the main trajectory or the backup trajectory and controls the vehicle to execute.

[0055] Compared with the prior art, the technical effects of the present application are:

[0056] The present invention provides an automobile automatic driving system and a behavioral safety method, which includes a parallel main channel and a backup channel, and the main channel and the backup channel are composed of a first environment perception module, a second environment perception module, an environment fusion module, a trajectory generation module, a backup trajectory generation module, a safety verification module, a state switching module, and a trajectory execution module. The present invention uses the first environment perception module and the second environment perception module as the environment perception of the main channel, thereby improving the robustness of the entire system to scenarios where the first environment perception module fails or has insufficient functions. The backup trajectory generation module is combined with the second environment perception module to output the backup trajectory, which facilitates the selection of the state switching module and improves the reliability of the system. The environment fusion module is used as the output of the first environment perception module and the second environment perception module in the main channel, and is output to the trajectory generation module of the main channel, thereby improving the environmental perception result. The main trajectory and the backup trajectory are verified by the safety verification module, without the need to correct the trajectory output, thereby reducing the possibility that safety verification and trajectory correction will affect the overall availability of the system. The state switching module is used to switch the trajectory output to the downstream, thereby achieving trajectory output redundancy and improving the reliability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 Schematic diagram of the system structure of the present invention;

[0058] Figure 2 Schematic diagram of the method of the present invention. DETAILED DESCRIPTION

[0059] The present invention is described in detail below with reference to the accompanying drawings:

[0060] In one embodiment of the present invention, an automatic driving system for a car is disclosed, referring to Figure 1 As shown, the system includes a parallel main channel and a backup channel; wherein, the main channel and the backup channel are composed of a first environment perception module, a second environment perception module, an environment fusion module, a trajectory generation module, a backup trajectory generation module, a security verification module, a state switching module, and a trajectory execution module.

[0061] Specifically, in this embodiment, the main channel is composed of a first environment perception module, a second environment perception module, an environment fusion module, a trajectory generation module, a security verification module, and a state switching module; the backup channel is composed of a second environment perception module, a backup trajectory generation module, a security verification module, and a state switching module.

[0062] In the main channel, the first and second environment perception modules receive sensor data respectively and transmit the perception results to the environment fusion module. The trajectory generation module receives the output of the environment fusion module, generates a main trajectory, and outputs the main trajectory to the security verification module and the state switching module.

[0063] In the standby channel, the second environment perception module receives sensor data and transmits the perception result to the standby trajectory generation module, the standby trajectory generation module generates a standby trajectory according to the received perception result, and outputs the standby trajectory to the safety verification module and the state switching module;

[0064] The state switching module determines to adopt the main trajectory or the standby trajectory based on the verification result output by the safety verification module, and outputs the adopted trajectory; the trajectory execution module receives and executes the trajectory output by the state switching module.

[0065] The modules included in the above-mentioned main channel and standby channel are analyzed in detail as follows:

[0066] 1. Environment perception (first environment perception module, second environment perception module)

[0067] The first environment perception module and the second environment perception module perceive and identify objects, road information, etc. in the environment around the vehicle through raw sensor data.

[0068] Input: raw sensor data, including camera data, lidar data, millimeter wave radar data (image, point cloud, radar data, etc.);

[0069] Output: perception result (object detection result), including the position, speed, category of objects in the environment, and road information, etc.

[0070] Implementation of the first environment perception module and the second environment perception module:

[0071] The first environment perception module processes sensor data based on a deep learning model to obtain a perception result;

[0072] The second environment perception module processes sensor data based on a traditional algorithm and a lightweight deep learning model to obtain a perception result.

[0073] 2. Environment fusion module

[0074] The environment fusion module fuses the perception results of the first environment perception module and the second environment perception module to provide more reliable environment perception results.

[0075] Input: perception results of the first environment perception module and the second environment perception module;

[0076] Output: fused perception result (fused scene understanding result).

[0077] The fusion method of the environment fusion module includes one of target level fusion algorithm, probability fusion method, and weighted voting mechanism.

[0078] The perception results of the first environment perception module and the second environment perception module are fused in front of the trajectory generation module by the environment fusion module, so as to reduce the state of misentry of the safe driving behavior caused by the false detection of the safety verification module that the first environment perception module is insufficient.

[0079] 3. The trajectory generation module

[0080] The trajectory generation module plans a driving path of the vehicle based on the fused perception results, and defines the path as a main trajectory.

[0081] Input: fused perception results, vehicle navigation target

[0082] Output: planned driving path of the vehicle (including position, speed, acceleration, etc.).

[0083] The trajectory generation module adopts a hierarchical planning architecture, a path planning algorithm and a speed planning algorithm to generate the main trajectory.

[0084] 4. The backup trajectory generation module

[0085] The backup trajectory generation module plans a driving path of the vehicle based on the perception results of the second environment perception module, and defines the path as a backup trajectory.

[0086] Input: perception results of the second environment perception module

[0087] Output: planned driving path of the vehicle (including position, speed, acceleration, etc.).

[0088] The backup trajectory generation module adopts a pure tracking algorithm, a safety distance setting and a preset emergency operation to generate the backup trajectory.

[0089] 5. The safety verification module

[0090] The safety verification module is used to verify the safety of the main trajectory and the backup trajectory. Unlike the "safety verification and trajectory correction module" in the traditional scheme, the safety verification module does not take charge of trajectory correction in this embodiment, so as to reduce the complexity of the module, improve the reliability of verification and reduce the false triggering caused by inappropriate preset rules. By verifying the main trajectory and the backup trajectory at the same time, the safety verification module provides reliable decision basis for the state switching module.

[0091] Input: main trajectory generated by the trajectory generation module, backup trajectory generated by the backup trajectory generation module, perception results of the first environment perception module and the second environment perception module

[0092] Output: safety verification result of the main trajectory and the backup trajectory.

[0093] The security verification mode of the security verification module is one or more of rule-based collision detection, space-time security corridor verification, and probability risk assessment.

[0094] It should be noted that the security verification module simultaneously receives the main trajectory generated by the main channel and the backup trajectory generated by the backup channel, reduces the false triggering caused by the unsuitable preset rules, and when the security verification module verifies that the main channel does not pass, the vehicle is automatically switched to the backup channel by default, thereby ensuring the safe driving of the vehicle.

[0095] 6. State switching module

[0096] The state switching module selects the trajectory to be finally executed according to the security verification result.

[0097] Input: main trajectory, backup trajectory, security verification result;

[0098] Output: final execution trajectory.

[0099] The selection process of the state switching module includes a normal working mode, a condition for triggering switching, switching logic, and special situation processing.

[0100] 7. Trajectory execution module

[0101] The trajectory execution module executes the trajectory selected by the state switching module, and realizes the movement of the vehicle according to the trajectory through a control instruction.

[0102] Input: final execution trajectory;

[0103] Output: control instruction (steering, acceleration, braking, etc.).

[0104] The implementation of the trajectory execution module adopts a lateral controller, a longitudinal controller, and an execution monitoring mechanism.

[0105] Through the above content, the automatic driving system of the present application is disclosed, which has the following advantages compared with the prior art:

[0106] 1. The first environment perception module and the second environment perception module are used as the environment perception of the main channel, which improves the robustness of the whole system to the failure and insufficient function of the first environment perception module.

[0107] 2. The backup trajectory generation module is combined with the second environment perception module to output the backup trajectory, which facilitates the selection of the state switching module and improves the reliability of the system.

[0108] 3. The environment fusion module is used to fuse the outputs of the first environment perception module and the second environment perception module in the main channel, and outputs to the trajectory generation module of the main channel, thereby improving the environment perception result.

[0109] 4. The application verifies the main trajectory and the backup trajectory through the safety verification module, without correcting the output of the trajectory, thereby reducing the possibility of affecting the overall availability of the system by safety verification and trajectory correction;

[0110] 5. The application adopts the state switching module to switch the trajectory output to the downstream, thereby realizing trajectory output redundancy and improving the reliability of the system.

[0111] In another embodiment of the application, a vehicle automatic driving behavior safety method is disclosed, as shown in the figure, the method is based on the above-mentioned automatic driving system, including a main trajectory generation process, a backup trajectory generation process, and a main trajectory and backup trajectory control process. Figure 2

[0112] The main trajectory generation process is realized based on the main channel and includes the following steps:

[0113] S11: The first environment perception module and the second environment perception module respectively receive sensor data and obtain perception results;

[0114] S12: The first environment perception module and the second environment perception module output the perception results to the environment fusion module, fuse the perception results of the first environment perception module and the perception results of the second environment perception module, and obtain the fused perception results;

[0115] S13: The environment fusion module outputs the fused perception results to the trajectory generation module for main trajectory generation;

[0116] S14: The trajectory generation module outputs the main trajectory to the safety verification module and the state switching module to verify the safety of the main trajectory.

[0117] The backup trajectory generation process is realized based on the backup channel and includes the following steps:

[0118] S21: The second environment perception module receives sensor data and obtains perception results;

[0119] S22: The second environment perception module outputs the perception results to the backup trajectory generation module for backup trajectory generation;

[0120] S23: The trajectory generation module outputs the backup trajectory to the safety verification module and the state switching module to verify the safety of the backup trajectory;

[0121] The main trajectory and backup trajectory control process includes the following steps:

[0122] S31: The safety verification module outputs the verification results of the main trajectory and the backup trajectory, the main trajectory, and the backup trajectory to the state switching module;

[0123] ​S32: The state switching module selects the main trajectory or the backup trajectory according to the verification result and outputs the selected trajectory to the trajectory execution module;

[0124] S33: The trajectory execution module receives the main trajectory or the backup trajectory and controls the vehicle to execute.

[0125] Specifically, the embodiment details the steps of the behavior safety method implemented by the modules of the automatic driving system as follows:

[0126] In the steps S11 and S21, the first and second environment perception modules perceive the input raw sensor data and output the perception results.

[0127] The first environment perception module uses a deep learning model for high-precision perception. The perception process mainly includes data preprocessing, feature extraction, target detection and recognition, and result output, as follows:

[0128] Data preprocessing:

[0129] The input raw sensor data (such as images, point clouds, radar data, etc.) are preprocessed, including cleaning, denoising, calibration, etc., to ensure the accuracy and consistency of the data.

[0130] Feature extraction:

[0131] The preprocessed data are input into the deep learning model. The model automatically extracts features from the data, including edges, textures, shapes, depths, etc., through convolution layers, pooling layers, etc.

[0132] Target detection and recognition:

[0133] Based on the extracted features, the deep learning model detects objects and determines their positions, speeds, and category information.

[0134] Result output:

[0135] The deep learning model outputs the detection results in the form of object detection results, including the position, speed (if measurable), and category of each detected object.

[0136] The second environment perception module uses traditional algorithms and lightweight deep learning models for high-precision perception. The perception process mainly includes data preprocessing, feature extraction, target detection and preliminary recognition, deep learning and further recognition, result output and post-processing, as follows:

[0137] Data preprocessing:

[0138] Preprocessing of raw sensor data (such as images, point clouds, radar data, etc.) including cleaning, denoising, calibration, etc. to ensure data accuracy and consistency;

[0139] Feature extraction:

[0140] Preprocessed data is used to extract key features from the data, including edges, textures, shapes, depths, etc., using traditional algorithms (such as edge detection, filtering in image processing) and simplified deep learning models (such as lightweight CNN networks such as MobileNets).

[0141] Target detection and preliminary identification:

[0142] Determine the location of the target (such as the bounding box) and preliminarily judge the category of the target (such as pedestrians, vehicles, obstacles, etc.) through traditional algorithms.

[0143] Deep learning and further identification:

[0144] For targets that require higher accuracy identification, the first environmental perception module inputs the above identification results into a more complex deep learning model for further identification, which includes deeper neural networks, more complex feature extraction layers, etc.

[0145] Result output and post-processing:

[0146] The detection results of the second environmental perception are output in the form of object detection results, including the location, speed (if measurable), category, etc. of each detected object.

[0147] In the above step S12, the environmental fusion module fuses the perception results of the first environmental perception module and the second environmental perception module, and outputs the fused perception results.

[0148] In this embodiment, the fusion method of the environmental fusion module includes but is not limited to target-level fusion algorithm, probability fusion method, weighted voting mechanism.

[0149] The fusion process of the target-level fusion algorithm is as follows: the environmental fusion module receives the perception results from the first environmental perception module and the second environmental perception module; uses a target tracking algorithm to associate the perception results of the first environmental perception module and the second environmental perception module; according to the association result, makes a fusion decision, for the associated objects, takes the reliability and accuracy of their position, speed, etc. Weighted fusion; output the fused perception results, including the position, speed, category of the object.

[0150] The fusion process of the probability fusion method (Bayesian fusion) is as follows: the environment fusion module receives the perception results from the first environment perception module and the second environment perception module; the probabilities of the presence of objects are calculated for the perception results of the first environment perception module and the second environment perception module; the probability results of the first environment perception module and the second environment perception module are fused by using the Bayes theorem or a related probability fusion method to obtain the posterior probability of the presence of the objects; whether the objects are included in the fused perception results (usually the objects with a probability exceeding a set threshold) is determined according to the fused probability results, and the position, speed, and category information of the objects are determined, and the fused perception results are output.

[0151] The fusion process of the weighted voting mechanism is as follows: the environment fusion module receives the perception results from the first environment perception module and the second environment perception module; the perception results of the first environment perception module and the second environment perception module are assigned weights based on historical performance evaluation, real-time sensor state, or other related factors; the perception results of each object are weighted and voted, and whether the object is included in the fused perception results is determined according to the support degree of the weighted voting; the fused perception results are determined and output according to the results of the weighted voting.

[0152] In the step S13, the trajectory generation module plans a driving trajectory based on the fused perception results, including the following steps:

[0153] The trajectory generation module receives the fused perception results and the navigation target, and pre-processes the input data, including data cleaning, verification, formatting, coordinate conversion, time synchronization, and the like;

[0154] An optimal or feasible path from the current position to the navigation target is searched in the environment map by using a path planning algorithm; the length, safety, feasibility, and the like of the searched multiple paths are evaluated, and an optimal path is selected as the driving path of the vehicle;

[0155] When the path is selected, a smooth speed curve is generated by using a speed planning algorithm (such as dynamic programming or quadratic programming) according to road conditions, traffic rules, obstacle distribution, and the like;

[0156] Based on the speed curve, the acceleration and deceleration requirements of the vehicle at each position point are calculated to ensure that the vehicle can smoothly drive according to the planned speed curve;

[0157] The driving path and the speed curve obtained above are combined to generate a detailed motion trajectory, and the generated trajectory is optimized, and the optimized trajectory is output as the main trajectory.

[0158] In the step S22, the backup trajectory generation module plans a driving trajectory based on the perception results of the second environment perception module, including the following steps:

[0159] The backup trajectory generation module receives the perception results of the second environment perception module, and adopts a simplified planning algorithm, such as a pure pursuit algorithm, to follow a path, directly calculates a path to be followed by the vehicle based on the current environment perception results and a dynamic model of the vehicle, to ensure fast response and calculation stability;

[0160] Speed planning is performed, and a safety distance is set, and the expression is:

[0161] d_safe = v * t_response + v² / (2*a_max) + margin;

[0162] wherein v is the current speed, t_response is the system response time, a_max is the maximum deceleration, and margin is an additional safety margin (larger than the main channel);

[0163] According to possible emergency situations of the vehicle, a series of emergency operation schemes are preset, including but not limited to an emergency braking scheme, an emergency avoidance scheme, and a safe parking plan; the emergency braking scheme: when it is detected that the distance to the front obstacle is less than the safety distance, a braking with the maximum allowed deceleration is performed; the emergency avoidance scheme: when there is an available avoidance space, a predefined avoidance trajectory is performed; the safe parking plan: when the system cannot ensure the safety of continuous driving, a shoulder or safe lane safe parking procedure is performed;

[0164] The results of the above steps are integrated to generate a backup safety trajectory, which is output as a backup trajectory.

[0165] In the step S14 and the step S23, the safety verification module verifies the safety of the main trajectory and the backup trajectory based on the input main trajectory, the backup trajectory, and the environment perception results, and outputs a safety verification result. Specifically, the process of the safety verification module includes the following steps:

[0166] The safety verification module receives the main trajectory, the backup trajectory, and the environment perception results;

[0167] For the received information, collision detection is performed based on predefined rules, including but not limited to a minimum safety distance between the vehicle and the obstacle, a speed limit, a traffic rule, etc.; by comparing the positions of the vehicle on the main trajectory and the backup trajectory with the positions of the obstacles in the environment, it is determined whether there is a potential collision risk;

[0168] For the trajectory after collision detection, a time-space safety corridor verification is performed to check whether the main trajectory and the backup trajectory are always kept within the time-space safety corridor, to ensure the safety of the vehicle during driving;

[0169] For the trajectory after the verification of the space-time safety corridor, a probability risk assessment is performed, and based on the uncertainty factors in the environment (including weather conditions, behaviors of traffic participants, etc.), the safety of the main trajectory and the backup trajectory is assessed; through calculation of collision probability, accident severity, etc., a more comprehensive safety assessment result is obtained;

[0170] Based on the above verification and assessment, the safety verification module outputs the safety verification result of the main trajectory and the backup trajectory, including the safety level of the trajectory, potential risk points, suggested risk avoidance measures, etc.

[0171] In the above step S32, the state switching module outputs the final execution trajectory based on the verification result of the main trajectory and the backup trajectory. Specifically, the switching logic of the state switching module includes the following:

[0172] In the normal working mode, the main trajectory output by the main channel is used by default, and the verification result of the safety verification module is continuously monitored;

[0173] The conditions for triggering switching include: when the safety verification module determines that the main trajectory has a safety risk, when the verification result of the safety verification module cannot be obtained, and when it is detected that the first environmental perception module of the main channel has insufficient function or failure;

[0174] The switching logic includes immediate switching, smooth switching, and conservative strategy; in immediate switching, the backup trajectory is switched to immediately when a safety risk is detected; in smooth switching, trajectory smooth transition is adopted in non-emergency situations; in the conservative strategy, the backup trajectory is switched to by default when the state is uncertain;

[0175] Special case processing includes: when the first environmental perception module of the main channel has insufficient function, switching to the backup channel to ensure safe driving behavior and monitor the output quality of the environmental fusion module; when the first environmental perception module of the main channel fails, detecting that the trajectory generation module outputs abnormally, and automatically switching to the backup trajectory; when the safety verification module does not respond, switching to the backup trajectory by default and starting a fault diagnosis program.

[0176] In the above step S33, the trajectory execution module controls the vehicle based on the execution trajectory finally determined by the state switching module, specifically, generates specific control instructions according to the trajectory to control the vehicle to travel along the predetermined trajectory, including the following steps:

[0177] The trajectory execution module receives the execution trajectory finally determined by the state switching module, including the main trajectory or the backup trajectory;

[0178] The received trajectory is parsed to extract key information such as target position, target speed, target acceleration, etc., and the variation law of the above parameters over time;

[0179] A lateral controller (such as a PID controller, MPC controller, etc.) is used to generate steering control instructions according to the lateral information of the trajectory, which includes but is not limited to the curvature of the path and the deviation of the current position of the vehicle from the path.

[0180] A longitudinal controller is used to generate acceleration or braking control instructions according to the longitudinal information of the trajectory, which includes but is not limited to the difference between the target speed and the current speed.

[0181] The longitudinal controller may involve speed tracking, distance keeping, etc. strategies to ensure that the vehicle can smoothly follow the predetermined speed changes.

[0182] The control instructions generated by the lateral controller and the longitudinal controller are integrated to form the final control command, which is output to the actuators of the vehicle, such as the steering system, the throttle and brake system, to realize real-time control of the vehicle.

[0183] At the same time, a monitoring mechanism is implemented for the vehicle, and if a deviation is found to be too large or an abnormal situation is found, timely adjustment or triggering of an emergency handling mechanism is performed, and the state information during execution is fed back to the upper control system for further optimization and decision-making, and real-time detection of the difference between the actual response of the vehicle and the control instructions.

[0184] Through the above, the behavior safety method of the present application is disclosed, which constructs two environmental perception algorithm models, the model of the first environmental perception module of the main channel is the main perception algorithm model, which has complete functions and balances the false detection rate and the missed detection rate; the model of the second environmental perception module of the standby channel is the secondary perception algorithm model, which may not have complete functions, but should ensure the detection rate of safety risk objects; the construction and training of the second environmental perception module ensure that it has more robustness than the model of the first environmental perception module, therefore, the models of the first environmental perception module and the second environmental perception module have different construction and training methods. It should be noted that the construction and training methods of the first environmental perception module and the second environmental perception module in this embodiment are any method in the prior art, which is not within the scope of protection of this embodiment.

[0185] The above embodiments are only used to illustrate the technical solutions of the present application and not to limit it, and other modifications or equivalent replacements of the technical solutions of the present application made by those skilled in the art should be covered within the scope of the claims of the present application, as long as they do not deviate from the spirit and scope of the technical solutions of the present application.

Claims

1. An automatic driving system for a car, characterized by: The system includes a parallel main channel and a backup channel; wherein the main channel and the backup channel are composed of a first environment perception module, a second environment perception module, an environment fusion module, a trajectory generation module, a backup trajectory generation module, a safety verification module, a state switching module, and a trajectory execution module; The main channel is composed of a first environment perception module, a second environment perception module, an environment fusion module, a trajectory generation module, a security verification module, and a state switching module; The first environment perception module and the second environment perception module respectively receive sensor data and transmit the perception results to the environment fusion module. The trajectory generation module receives the output of the environment fusion module, generates a main trajectory, and outputs the main trajectory to the security verification module and the state switching module. The backup channel consists of a second environment perception module, a backup trajectory generation module, a safety verification module, and a state switching module; The second environment perception module receives sensor data and transmits the perception results to the backup trajectory generation module. The backup trajectory generation module generates a backup trajectory based on the received perception results and outputs the backup trajectory to the security verification module and the state switching module. The state switching module decides whether to adopt the primary trajectory or the backup trajectory based on the verification result output by the safety verification module, and outputs the adopted trajectory; The trajectory execution module receives and executes the trajectory output by the state switching module.

2. The automatic driving system for a vehicle according to claim 1, characterized in that: The input data of the first and second environmental perception modules are raw sensor data, including camera data, lidar data, and millimeter-wave radar data, and the output data are perception results, including the position, speed, category, and road information of objects in the environment; The first environment perception module processes sensor data based on a deep learning model to obtain a perception result; The second environmental perception module processes sensor data based on traditional algorithms and lightweight deep learning models to obtain perception results.

3. The automatic driving system for a vehicle according to claim 1, characterized in that: The environment fusion module takes the perception results of the first environment perception module and the second environment perception module as input to fuse them, obtains a fused perception result, and outputs the fused perception result; The fusion mode of the environment fusion module includes one of a target level fusion algorithm, a probabilistic fusion method, and a weighted voting mechanism.

4. The automatic driving system for a vehicle according to claim 1, characterized in that: The trajectory generation module takes the fused perception results output by the environment fusion module and the vehicle navigation target as input, outputs the planned driving path of the vehicle, and defines the path as the main trajectory; The trajectory generation module uses a hierarchical planning architecture, a path planning algorithm, and a speed planning algorithm to generate the main trajectory.

5. The automatic driving system for a vehicle according to claim 1, characterized in that: The backup trajectory generation module takes the perception result of the second environment perception module as input, outputs the planned driving path of the vehicle, and defines the path as the backup trajectory; The backup trajectory generation module uses a pure tracking algorithm, a safety distance setting, and a preset emergency operation to generate a backup trajectory.

6. The automatic driving system for a vehicle according to claim 1, characterized in that: The safety verification module takes the main trajectory generated by the trajectory generation module, the backup trajectory generated by the backup trajectory generation module, and the perception result as input, performs safety verification, and outputs the verification results of the main trajectory and the backup trajectory; The safety verification method of the safety verification module is one or more of rule-based collision detection, spatiotemporal safety corridor verification, and probabilistic risk assessment.

7. The automatic driving system for a vehicle according to claim 1, characterized in that: The state switching module selects the final execution trajectory according to the verification result of the safety verification module; the input of the state switching module includes the main trajectory, the backup trajectory, and the safety verification result, and outputs the final execution trajectory; The selection process of the state switching module includes normal working mode, conditions for triggering switching, switching logic, and special case processing.

8. The automatic driving system for a vehicle according to claim 1, characterized in that: The trajectory execution module realizes the movement of the vehicle according to the trajectory through control instructions based on the output trajectory of the state switching module; The trajectory execution module is implemented by using a lateral controller, a longitudinal controller, and an execution monitoring mechanism.

9. A method for ensuring the safety of autonomous driving behavior of a vehicle according to any one of claims 1 to 8, characterized in that: The method includes a main trajectory generation process, a backup trajectory generation process, and a main trajectory and backup trajectory control process; The main trajectory generation process is based on the main channel and includes the following steps: S11: The first environment perception module and the second environment perception module respectively receive sensor data and obtain perception results; S12: The first environment perception module and the second environment perception module output perception results to the environment fusion module, which fuses the perception results of the first environment perception module with the perception results of the second environment perception module to obtain a fused perception result; S13: The environment fusion module outputs the fused perception results to the trajectory generation module to generate the main trajectory; S14: The trajectory generation module outputs the main trajectory to the safety verification module and the state switching module to verify the safety of the main trajectory; The backup trajectory generation process is based on the backup channel and includes the following steps: S21: The second environment perception module receives sensor data and obtains a perception result; S22: The second environment perception module outputs the perception result to the backup trajectory generation module to generate a backup trajectory; S23: The trajectory generation module outputs the backup trajectory to the safety verification module and the state switching module to verify the safety of the backup trajectory; The main trajectory and backup trajectory control process includes the following steps: S31: The safety verification module outputs the verification results of the main trajectory and the backup trajectory, the main trajectory, and the backup trajectory to the state switching module; S32: The state switching module selects to output the main trajectory or the backup trajectory to the trajectory execution module according to the verification result; S33: The trajectory execution module receives the main trajectory or the backup trajectory and controls the vehicle to execute it.

Images