Intelligent vehicle limit safety system under extreme working condition

Abstract

The application discloses an intelligent vehicle limit safety system under extreme working conditions, and the system comprises a vehicle sensing system supporting a limit safety function, which is used for collecting environmental information of the vehicle under extreme working conditions and transmitting the environmental information to an automatic driving system and a high-level auxiliary driving system in real time; a limit safety function decision module of the automatic driving system and the high-level auxiliary driving system, which is used for acquiring the environmental information and receiving input information based on a data-driven reinforcement learning strategy, vehicle state feedback and execution information in an unknown environment; and a limit safety function execution module, which is used for acquiring decision information and outputting the vehicle state and the execution information. The application can be used for operations under extreme working conditions, which include but are not limited to operations related to dynamic critical stability or limit motion capability, so as to fully utilize the motion potential of each system of the intelligent vehicle and improve the emergency processing capability and backup level of the intelligent vehicle under extreme working conditions.

Description

Technical Field

[0001] The present invention relates to the technical field of automobile active safety, and in particular to an extreme safety system for intelligent vehicles under extreme working conditions. Background Art

[0002] Extreme operating conditions generally refer to critical environments with a risk of collision or conditions where the vehicle's own condition has seriously deteriorated. Relevant scenarios include sudden forward accidents, high-speed narrow bends, and critical vehicle instability. Under such operating conditions, the vehicle's reaction speed and decision-making intelligence level are particularly critical. The control concept of the active safety system currently installed in vehicles is mainly to limit the vehicle's driving state to a linear and steady-state range to avoid vehicle instability when the wheels reach their dynamic limits. From the perspective of vehicle controllability, the current active safety functions are too conservative and cannot fully utilize the vehicle's dynamic control potential under extreme conditions.

[0003] At present, the design of safety functions for extreme working conditions is still in its early stages. Active safety functions at the product level, such as automatic emergency braking, target relatively single scenarios and have simple control strategies. Functional design at the research level is relatively fragmented, with limited utilization of the vehicle's dynamic control potential. There is also a lack of a unified framework to integrate different technical methods, which limits the synergistic effect of vehicle execution systems in actual applications.

[0004] Currently, theoretical approaches to extreme safety functions for extreme scenarios have yet to be established, and the triggering criteria, decision-making planning, and control methods for these functions are still unsystematic. Regarding triggering criteria, there is no unified triggering system that comprehensively considers collision avoidance, cornering, and dynamic stability requirements. This leads to inconsistencies and limitations in the application of triggering criteria in different extreme situations. Regarding decision-making planning, there is a lack of an evaluation system for decision-making in a wide range of extreme scenarios. Regarding control methods, there is a lack of a unified framework to integrate the advantages of various technical approaches, resulting in limited synergy between current technologies in practical applications.

[0005] In response to the above problems, there is an urgent need to establish an extreme safety function, architecture and system with unified triggering standards, a complete evaluation system and a collaborative control solution. Summary of the Invention

[0006] The present invention aims to solve one of the technical problems in the related art at least to a certain extent.

[0007] To this end, the present invention proposes an extreme safety system for intelligent vehicles under extreme working conditions, which performs operations including but not limited to those involving critical dynamic stability or extreme motion capabilities under extreme working conditions, so as to fully utilize the motion potential of various systems of the intelligent vehicle and improve the emergency handling capability and backup level of the intelligent vehicle under extreme working conditions.

[0008] To achieve the above objectives, the present invention provides, on one hand, an extreme safety system for intelligent vehicles under extreme operating conditions, comprising:

[0009] The vehicle sensing system, which supports extreme safety functions, collects environmental information under extreme vehicle operating conditions and transmits this information in real time to the autonomous driving system and advanced driver assistance systems via Ethernet or CAN communication.

[0010] The autonomous driving system and high-level assisted driving system include an extreme safety function decision module and an extreme safety function execution module, wherein;

[0011] The extreme safety function decision module includes an extreme collision avoidance judgment module, an extreme cornering judgment module, an extreme vehicle condition identification module, an extreme state identification module, and a safety margin monitoring module; the extreme safety function decision module is used to obtain environmental information and receive input information based on a data-driven reinforcement learning strategy in an unknown environment, as well as vehicle status and execution information fed back by the extreme safety function execution module;

[0012] The extreme safety function execution module is used to obtain the decision information sent by the extreme safety function decision module and output the vehicle status and execution information.

[0013] The extreme safety system for intelligent vehicles under extreme operating conditions according to the embodiment of the present invention may also have the following additional technical features:

[0014] In one embodiment of the present invention, the vehicle sensing system supporting the extreme safety function includes a camera, a millimeter-wave radar, and a combined inertial navigation unit (IMU).

[0015] In one embodiment of the present invention, the extreme safety function decision module is further configured to:

[0016] The target trajectory is solved by an extreme safety decision solver based on the evaluation results and using an iterative optimization or dynamic optimization method; wherein the evaluation results include the characteristics of the extreme scenario, and the evaluation results are obtained by evaluating the vehicle's risk avoidance response in the extreme scenario in combination with the direct risk of the scenario, the expected risk of the scenario, the operational risk avoidance level, and the degree of operational limit;

[0017] All target trajectories of the decision solution that match the current extreme scenario are passed to the extreme safety function execution module.

[0018] In one embodiment of the present invention, the system is further configured to utilize a fusion control mechanism to fuse hierarchical decision control and data-driven decision control; wherein,

[0019] The fusion control mechanism is designed for situations where hierarchical decision control and data-driven decision control have the same objectives: decisions are deployed in a data generation environment and trajectories are generated, replacing the trajectory information output by hierarchical decision making with trajectories; a model-based trajectory tracker is used to solve and generate closed-loop inputs. Directly generate feedforward input using data-driven The two are added or weighted to generate the final input to be executed:

[0020]

[0021] The fusion control mechanism is aimed at the situation where the hierarchical decision control and data-driven decision control have different control objectives: by comparing the target gap between data-driven decision and hierarchical decision, the data-driven strategy directly generates action information a Data , track the hierarchical decision target trajectory to obtain action information a model ,Based on the target difference, the final input to be executed is generated, and the generation methods include neural network fusion and weighted fusion:

[0022] a t =Diff target (a Data ,a model ).

[0023] In one embodiment of the present invention, for the execution input output by the fusion control mechanism, the four-wheel drive and rear-wheel drive are combined in driving to control the drive slip and expand the vehicle's extreme driving control boundary; the friction brake and motor brake are combined in braking to provide a backup extreme braking capacity guarantee; the active steering and differential steering are combined in steering to provide maximum posture adjustment capability for the extreme safety function; each actuator of the vehicle executes the target input and performs extreme safety maneuvers.

[0024] In one embodiment of the present invention, the environmental information is a sharp bend environment, and the goal of the extreme safety decision module is to ensure comprehensive safety of extreme drift cornering; the evaluation indicators related to extreme cornering performance and safety include immediate evaluation and final evaluation: the immediate cornering performance indicators include: the evaluation item r for tracking the suboptimal trajectory of the scene p :

[0025]

[0026] Among them, k pl and k pv is a negative parameter; l is the lateral deviation of the vehicle relative to the center of the curve in the Frenet coordinate system, v is the current vehicle speed; l ref (s) is the lateral displacement of the suboptimal trajectory of the formation at the current bend center; is the maximum vehicle speed under the specific stability constraints under the current adhesion.

[0027] In one embodiment of the present invention, the instant evaluation item r for rewarding high sideslip angles is β , encouraging the vehicle to turn with a higher center-of-mass slip angle:

[0028]

[0029] Among them, k β is a negative number, v x is the longitudinal velocity in the vehicle coordinate system, v y is the lateral speed in the vehicle coordinate system, β is the sideslip angle of the vehicle’s center of mass;

[0030] The final reward item r is used to reward the comprehensive performance of the extreme safety function t , encourages safe and fast cornering:

[0031] r t =(1-χ)·k t1 +k t2 ·χ·(t f -t ref )

[0032] Where χ represents the parameter of the vehicle's final state: χ = 1 means the vehicle has safely completed the turning task, χ = 0 means an unsafe event has occurred, t f Indicates the time required for the vehicle to reach the terminal state, t ref represents the total travel time of the pre-optimized trajectory; the constant k t1 and k t2 are negative and positive values, respectively, used to penalize unsafe terminal states and reward the shortest possible extreme turning time.

[0033] In one embodiment of the present invention, the extreme safety decision solver adopts an end-to-end iterative solver based on reinforcement learning, and the output is directly action information; the iterative method is a Critic and Actor deep neural network, and the Critic network parameters are trained according to minimizing the time difference loss function; the Actor network parameters are trained by maximizing the value function.

[0034] In one embodiment of the present invention, the data acquisition environment used in the extreme safety decision solver is the Carsim platform; the control strategy based on reinforcement learning is deployed to the corresponding scenario in the data acquisition environment to obtain the optimal trajectory T in the current scenario. p , the trajectory is used as the target trajectory information of the extreme safety function execution module:

[0035] T p ={S i |simulate(P v ,V c ,S ini,π),i=1,2,3,…,n}

[0036] Among them, S represents the vehicle state along the estimated trajectory, S ini Indicates the vehicle status when entering a curve.

[0037] In one embodiment of the present invention, since the estimated target trajectory is used to generate feedback input, it is necessary to set T p Convert to Cartesian coordinates Tracking the trajectory using a proportional-integral-derivative controller Get supplementary input a PID ; The end-to-end action output by reinforcement learning will serve as an important feedforward reference and be recorded as a RL Hierarchical decision-making has the same goal as data-driven decision-making, and uses direct addition to obtain the execution input:

[0038] a t =a RL +a PID .

[0039] The intelligent vehicle extreme safety system for extreme operating conditions in this embodiment of the present invention integrates traditional friction braking with modern motor braking technology. This system not only provides precise braking force but also increases braking redundancy, ensuring high safety in extreme braking situations. In terms of steering technology, this embodiment combines active steering and differential steering technology, significantly enhancing the vehicle's maneuverability and stability in emergency situations and providing maximum posture adjustment capabilities for extreme safety functions.

[0040] Additional aspects and advantages of the present invention will be set forth in part in the description which follows and, in part, will be obvious from the description which follows, or may be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0042] Figure 1 is a structural diagram of an extreme safety system for an intelligent vehicle under extreme working conditions according to an embodiment of the present invention;

[0043] Figure 2 is a deployment diagram according to an embodiment of the present invention;

[0044] Figure 3 2 is a deployment comparison trajectory diagram according to an embodiment of the present invention. DETAILED DESCRIPTION

[0045] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments of the present invention can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0046] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0047] The following describes an extreme safety system for an intelligent vehicle under extreme working conditions according to an embodiment of the present invention with reference to the accompanying drawings.

[0048] Figure 1 This is a structural diagram of the extreme safety system of the intelligent vehicle under extreme working conditions of the present invention. Figure 1 Shown, including:

[0049] The vehicle sensing system 10, which supports extreme safety functions, is used to collect environmental information under extreme vehicle operating conditions and transmit this information in real time to the autonomous driving system and advanced driver assistance system 20 via Ethernet or CAN communication;

[0050] The automatic driving system and high-level assisted driving system 20 includes an extreme safety function decision module 21 and an extreme safety function execution module 22, wherein;

[0051] The extreme safety function decision module 21 includes an extreme collision avoidance judgment module, an extreme cornering judgment module, an extreme vehicle condition identification module, an extreme state identification module, and a safety margin monitoring module. The extreme safety function decision module is used to obtain environmental information and receive input information based on a data-driven reinforcement learning strategy in an unknown environment, as well as vehicle status and execution information fed back by the extreme safety function execution module 22.

[0052] The extreme safety function execution module 22 is used to obtain the decision information sent by the extreme safety function decision module 21 and output the vehicle status and execution information.

[0053] To achieve the above objectives, the extreme safety system for intelligent vehicles under extreme working conditions of the present invention adopts the following technical solutions: extreme safety functions integrated in the advanced driver assistance system (ADAS) or automatic driving system of intelligent vehicles; modular extreme safety function decision and control architecture, including decision intervention models for extreme scenarios, extreme safety function optimizers and vehicle response evaluation systems; extreme operation execution methods that integrate the advantages of various solution technologies, including input fusion control mechanisms and cross-system collaborative input execution mechanisms.

[0054] For example, the proposed vehicle sensing system supporting extreme safety functions includes one or more of the following sensors, or a combination of sensors: cameras, millimeter-wave radar, and integrated inertial navigation. Given that extreme operating conditions often occur at high speeds, other sensors suitable for obstacle sensing at high speeds should also be included. This system transmits environmental information in real time to the extreme safety function decision-making and execution module via Ethernet or CAN communication.

[0055] For example, a proposed extreme safety function integrated into an intelligent vehicle's Advanced Driver Assistance System (ADAS) or Autonomous Driving System (ADS) is proposed, including but not limited to the use of maneuvers involving near-dynamic stability, such as drifting. This function can be integrated into the active safety module of the vehicle's Advanced Driver Assistance System (ADAS) or Autonomous Driving System (ADS), elevating its own priority when extreme safety functions intervene, taking over or blocking vehicle stability functions or related active safety features, including but not limited to anti-lock braking systems and electronic stability control.

[0056] For example, a module for determining when an extreme safety function should be engaged is proposed. Given that extreme driving conditions are relatively rare in driving scenarios, this module is the only one running full-time. It monitors the extreme collision avoidance requirements, cornering requirements, vehicle conditions, and dynamic stability of the driving conditions in real time, and comprehensively determines whether to trigger the extreme safety function.

[0057] For example, an evaluation system for intelligent vehicle risk avoidance responses in extreme scenarios has been constructed. Based on the characteristics of extreme scenarios, the system evaluates the vehicle's risk avoidance response in these scenarios by combining direct scenario risk, expected scenario risk, operational risk avoidance level, and operational limit.

[0058] For example, an extreme safety decision solver was constructed. Combined with the previously established extreme scenario vehicle avoidance response evaluation system, methods including but not limited to iterative optimization and dynamic optimization were employed to solve for the target trajectory or end-to-end input information. Notably, the decision module of the present invention is compatible with various model-based or data-driven solution methods. All decision solution outputs matching the current extreme scenario are passed to the extreme safety function execution module.

[0059] For example, a limit manipulation fusion mechanism is proposed that integrates the advantages of various solution technologies. Hierarchical decision-making generally has strong environmental adaptability and typically outputs target trajectory information. Data-driven decision-making control generally has good global adaptability and typically outputs direct action information. This invention adopts a fusion control mechanism that combines the advantages of these two technical approaches.

[0060] Fusion control mechanism, for the case where hierarchical decision-making and data-driven decision-making have the same goal: deploy the decision in the data generation environment and generate a trajectory, and use this trajectory to replace the trajectory information output by the hierarchical decision-making. Furthermore, a model-based trajectory tracker is used to solve and generate a closed-loop input. Directly generate feedforward input using data-driven The two are added or weighted to generate the final input to be executed.

[0061]

[0062] Fusion control mechanism, for the case where hierarchical decision-making and data-driven decision-making have different goals: comparing the goal gap between data-driven decision-making and hierarchical decision-making, the data-driven strategy directly generates action information a Data , track the hierarchical decision target trajectory to obtain action information a model ,Considering the target difference, the final input to be executed is generated, and the generation methods include but are not limited to neural network fusion, weighted fusion, etc.

[0063] a t =Diff target (a Data ,a model )

[0064] For example, a cross-system coordinated extreme maneuver execution is proposed. Based on the execution input output by the fusion control mechanism, the driving system combines four-wheel drive and rear-wheel drive to precisely control drive slip and expand the vehicle's extreme driving control boundaries. Braking combines friction braking and electric motor braking to provide precise and backup extreme braking capabilities. Steering combines active steering and differential steering to provide maximum posture adjustment capabilities for extreme safety functions. Furthermore, each vehicle actuator executes the target input and performs an extreme safety maneuver. The executed maneuver and the resulting scenario evolution are transmitted to the risk avoidance response evaluation system for iterative / dynamic safety decision-making.

[0065] Specifically, the extreme safety system for intelligent vehicles under extreme working conditions proposed by the present invention can improve the emergency response capability under extreme working conditions, such as Figure 1 As shown,

[0066] Module 10 presents the architecture of the vehicle sensor system that supports extreme safety features. This system consists of cameras, millimeter-wave radar, and an integrated inertial navigation unit (IMU). Data is transmitted through Ethernet and CAN FD integration with the autonomous driving system and advanced driver assistance systems.

[0067] Module 21 provides the architecture of the extreme safety function decision module, which consists of intervention judgment, comprehensive evaluation, and optimization solution modules. Module 21 obtains environmental information from module 10 and vehicle status and execution information from module 22.

[0068] Module 22 provides the architecture for executing extreme safety functions, consisting of multiple modules for data processing, input fusion, and coordinated execution. Module 22 obtains decision information from module 21. In this embodiment, module 21 receives input based on a data-driven reinforcement learning strategy in an unknown environment.

[0069] In this embodiment, module 10 detects a sharp curve with a diameter of 11 meters and sends relevant information to the extreme safety function decision module. The extreme cornering determination module generates an activation flag and transmits it to the extreme safety decision module. The extreme safety decision module considers the extreme cornering conditions, vehicle speed, and driver intent to trigger the extreme safety function, aiming to ensure comprehensive safety during extreme drift cornering.

[0070] After the extreme safety decision module intervenes, the vehicle stability function and other active safety functions such as collision avoidance are blocked.

[0071] It is understandable that this embodiment only lists the items in the evaluation system related to the embodiment scenario. Specifically, only the safety and performance evaluation indicators of the extreme cornering scenario are listed.

[0072] In this embodiment, the evaluation indicators related to extreme cornering performance and safety include immediate evaluation and final evaluation. The immediate cornering performance indicators include: evaluation items r for tracking the suboptimal trajectory of the scene p :

[0073]

[0074] Among them, k pl and k pv is a negative parameter; l is the lateral deviation of the vehicle relative to the center of the curve in the Frenet coordinate system, v is the current vehicle speed; l ref (s) is the lateral displacement of the suboptimal trajectory of the formation at the current bend center; is the maximum vehicle speed under the current stability constraint. This reward encourages driving along the pre-optimized path at the maximum speed allowed by the stability constraint, thereby improving the exploration efficiency during the iterative optimization process.

[0075] Instant evaluation item r used to reward high sideslip angle β , encouraging the vehicle to turn with a higher center-of-mass slip angle:

[0076]

[0077] Among them, k β is a negative number, v x is the longitudinal velocity in the vehicle coordinate system, v y is the lateral speed in the vehicle coordinate system, and β is the sideslip angle of the vehicle's center of mass.

[0078] The final reward item r is used to reward the comprehensive performance of the extreme safety function t , encourages safe and fast cornering:

[0079] r t =(1-χ)·k t1 +k t2 ·χ·(t f -t ref )

[0080] Here, χ represents the parameter of the vehicle's final state: χ = 1 means the vehicle has safely completed the turning task, while χ = 0 means an unsafe event has occurred, such as a collision with the track boundary or a rollover. f Indicates the time required for the vehicle to reach the terminal state, t ref represents the total travel time of the pre-optimized trajectory. t1 and k t2 are negative and positive values, respectively, used to penalize unsafe terminal states and reward the shortest possible extreme turning time.

[0081] In this embodiment, the extreme safety solver uses an end-to-end iterative solver based on reinforcement learning, and the output is directly action information. The constructed state space and action space are determined according to the actual situation of the deployed vehicle. Since common technologies in the field of reinforcement learning optimization are used, this embodiment will not elaborate on them in detail. In this embodiment, the iterative method is a Critic and Actor deep neural network. The Critic network parameters are trained by minimizing the time difference loss function; the Actor network parameters are trained by maximizing the value function. The above is only a specific example given in this embodiment. The strategies of imitation learning and reinforcement learning are both included in this embodiment.

[0082] In this embodiment, the data acquisition environment used in the extreme safety solver is the Carsim platform, where the vehicle parameters are common Class-C parameters, and the stability function interface is open. This embodiment is also compatible with self-built vehicle dynamics models or other dynamic response data acquisition platforms.

[0083] In this embodiment, the friction coefficient in the real environment is close to but not completely consistent with that of the data acquisition platform, and the inconsistency of tire types is introduced as an interference item.

[0084] In order to achieve precise cornering control, the control strategy based on reinforcement learning is deployed to the corresponding scenario in the data acquisition environment to obtain the optimal trajectory T in the current scenario. p , this trajectory serves as the target trajectory information of the extreme safety function execution module:

[0085] T p ={S i |simulate(Pv ,V c ,S ini ,π),i=1,2,3,…,n}

[0086] Where S represents the vehicle state along the estimated trajectory, S ini Represents the vehicle state when entering a curve. Note that since the estimated trajectory is used to generate feedback input, T p Convert to Cartesian coordinates The conversion details are omitted here.

[0087] This embodiment uses a proportional-integral-derivative controller to track the trajectory Get supplementary input a PID Please note that this embodiment includes all other methods for obtaining trajectory tracking input, including but not limited to linear quadratic Gaussian controllers and model predictive controllers. Specifically, the calculated input will be passed to the input fusion mechanism for further processing.

[0088] In this embodiment, the end-to-end action output by reinforcement learning will be used as an important feedforward reference and recorded as a RL This type of input usually has good global optimality and worrying environmental adaptability. Specifically, the input will be directly passed to the input fusion mechanism for further processing.

[0089] In this embodiment, since the hierarchical decision and data-driven decision have the same goal, a direct addition method is used to obtain the execution input:

[0090] a t =a RL +a PID

[0091] In this embodiment, the input to be executed a output by the fusion control mechanism t In terms of drive, this embodiment adopts a combination of four-wheel drive and rear-wheel drive to accurately control the slip of the drive wheels, thereby expanding the control boundary of the vehicle under extreme driving conditions. Through this drive mode, the vehicle can maintain optimal power output and stability under various complex road conditions. In terms of the braking system, the system integrates traditional friction braking and modern motor braking technology, which not only provides precise braking force, but also increases the redundancy of braking, ensuring high safety under extreme braking conditions. In terms of steering technology, this embodiment combines active steering system and differential steering technology, which greatly enhances the vehicle's controllability and stability in emergency situations, and provides maximum posture adjustment capabilities for extreme safety functions.

[0092] Furthermore, the vehicle's actuators execute extreme safety maneuvers based on the target inputs from the coordination mechanism. These maneuvers and the resulting scenario evolution are transmitted to a previously established risk avoidance response evaluation system, which assesses the vehicle's safety response in real time and makes iterative or dynamic safety decision adjustments. This process ensures that the vehicle can take the most appropriate risk avoidance measures in changing environments and emergencies, maximizing the intelligent vehicle's emergency response capabilities and backup capabilities in extreme conditions.

[0093] The performance of this embodiment in the extreme cornering scenario is as follows Figure 2 and Figure 3 shown. Figure 3 The comparison trajectory shown shows that the control effect of the strategy proposed in this embodiment is significantly better than that of the control using only a data-driven strategy or a feedback controller. Of the three compared strategies, only the solution of this embodiment can control the vehicle to negotiate a 180-degree U-turn with an 11-meter diameter at the extreme state of high sideslip angle.

[0094] The beneficial effects of the present invention are:

[0095] 1. Enhanced Emergency Response and Backup Capabilities: This invention integrates advanced dynamic control technologies, enabling intelligent vehicles to demonstrate superior emergency response capabilities under extreme operating conditions. This includes, but is not limited to, efficient response capabilities in emergency braking, extreme steering, or high-speed cornering. Furthermore, multiple backup mechanisms for the braking and driving systems enhance vehicle safety and reliability in extreme situations.

[0096] 2. Optimized Decision-Making and Control Precision: A modular extreme safety function decision-making and control architecture integrates an extreme scenario decision intervention model and an extreme safety function optimizer, providing precise decision-making support for a wide range of extreme scenarios. This enables the vehicle to make the most appropriate dynamic response in various extreme environments, significantly improving control accuracy and efficiency.

[0097] 3. Enhanced vehicle dynamics: Through advanced sensing systems and control algorithms, this invention enables the vehicle to operate near its dynamic limits, such as in high-speed cornering and emergency obstacle avoidance, fully utilizing the vehicle's kinetic potential. This not only enhances the vehicle's performance but also significantly expands its operating range and safety margins.

[0098] 4. Achieving High Integration and System Synergy: Through a fusion control mechanism that integrates the advantages of various solution technologies, this invention achieves cross-system coordinated execution. This synergy not only improves operational consistency and coordination, but also optimizes the responsiveness and efficiency of the entire vehicle system, especially when handling complex and changing driving environments.

[0099] According to the extreme safety system for intelligent vehicles under extreme working conditions of an embodiment of the present invention, operations including but not limited to those involving critical dynamic stability or extreme motion capabilities are performed under extreme working conditions to fully utilize the motion potential of various systems of the intelligent vehicle and enhance the emergency handling capability and backup level of the intelligent vehicle under extreme working conditions.

[0100] In the description of this specification, the reference terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" mean that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and features of different embodiments or examples without contradiction.

[0101] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present invention, "plurality" means at least two, such as two, three, etc., unless otherwise specifically defined.

Claims

1. An extreme safety system for intelligent vehicles under extreme working conditions, characterized by: include: The vehicle sensing system, which supports extreme safety functions, collects environmental information under extreme vehicle operating conditions and transmits this information in real time to the autonomous driving system and advanced driver assistance systems via Ethernet or CAN communication. The autonomous driving system and high-level assisted driving system include an extreme safety function decision module and an extreme safety function execution module, wherein; The extreme safety function decision module includes an extreme collision avoidance judgment module, an extreme cornering judgment module, an extreme vehicle condition identification module, an extreme state identification module, and a safety margin monitoring module; the extreme safety function decision module is used to obtain environmental information and receive input information based on a data-driven reinforcement learning strategy in an unknown environment, as well as vehicle status and execution information fed back by the extreme safety function execution module; The extreme safety function execution module is used to obtain the decision information sent by the extreme safety function decision module and output the vehicle status and execution information; The extreme safety function decision module is further used to: The target trajectory is solved by an extreme safety decision solver based on the evaluation results and using an iterative optimization or dynamic optimization method; wherein the evaluation results include the characteristics of the extreme scenario, and the evaluation results are obtained by evaluating the vehicle's risk avoidance response in the extreme scenario in combination with the direct risk of the scenario, the expected risk of the scenario, the operational risk avoidance level, and the degree of operational limit; Pass all target trajectories that match the current extreme scenario’s decision solution to the extreme safety function execution module; The system is also used to fuse hierarchical decision control and data-driven decision control using a fusion control mechanism; wherein, The fusion control mechanism is designed for situations where hierarchical decision control and data-driven decision control have the same objectives: decisions are deployed in a data generation environment and trajectories are generated, replacing the trajectory information output by hierarchical decision making with trajectories; a model-based trajectory tracker is used to solve and generate closed-loop inputs. , using data-driven direct generation of feedforward input , the two are added or weighted to generate the final input to be executed: The fusion control mechanism is aimed at the situation where the hierarchical decision control and data-driven decision control have different control objectives: by comparing the target gap between data-driven decision and hierarchical decision, the data-driven strategy directly generates action information. , track the hierarchical decision target trajectory to obtain action information ,Based on the target difference, the final input to be executed is generated, and the generation methods include neural network fusion and weighted fusion: 。 2. The system according to claim 1, wherein: The vehicle sensing system that supports extreme safety functions includes cameras, millimeter-wave radars, and combined inertial navigation units (IMUs).

3. The system according to claim 1, wherein: Based on the execution input of the fusion control mechanism output, the drive system combines four-wheel drive and rear-wheel drive to control drive slip and expand the vehicle's extreme drive control boundaries; the braking system combines friction braking and electric braking to provide backup extreme braking capacity; the steering system combines active steering and differential steering to provide maximum posture adjustment capabilities for extreme safety functions; Each actuator of the vehicle executes the target input and performs extreme safety maneuvers.

4. The system according to claim 1, wherein: The environmental information is a sharp bend environment, and the goal of the extreme safety decision module is to ensure comprehensive safety during extreme drift cornering. Evaluation indicators involving extreme cornering performance and safety include immediate evaluation and final evaluation: Instant cornering performance indicators include: Evaluation items for tracking suboptimal trajectories in scenarios : in, and is a negative parameter; is the lateral deviation of the vehicle relative to the center of the curve in the Frenet coordinate system, is the current vehicle speed; is the lateral displacement of the suboptimal trajectory of the formation at the current bend center; is the maximum vehicle speed under the specific stability constraints under the current adhesion.

5. The system according to claim 4, characterized in that Instant evaluation item used to reward high sideslip angles , encouraging the vehicle to turn with a higher center-of-mass slip angle: in, is a negative number, is the longitudinal speed in the vehicle coordinate system, is the lateral speed in the vehicle coordinate system, is the vehicle's sideslip angle; Final reward item for rewarding comprehensive performance of extreme safety functions , encourages safe and fast cornering: in, Parameters representing the final state of the vehicle: =1 means the vehicle has completed the turning task safely. =0 indicates an unsafe event occurs. Indicates the time required for the vehicle to reach the destination state. represents the total travel time of the pre-optimized trajectory; constant and are negative and positive values, respectively, used to penalize unsafe terminal states and reward the shortest possible extreme turning time.

6. The system according to claim 1, wherein: The extreme safety decision solver adopts an end-to-end iterative solver based on reinforcement learning, and the output is directly action information; the iterative method is and Deep neural network, trained by minimizing the temporal difference loss function Network parameters; trained by maximizing the value function Network parameters.

7. The system according to claim 1, wherein: The data acquisition environment used in the extreme safety decision solver is the Carsim platform; the control strategy based on reinforcement learning is deployed to the corresponding scenario in the data acquisition environment to obtain the optimal trajectory in the current scenario. , the trajectory is used as the target trajectory information of the extreme safety function execution module: in, represents the vehicle state along the estimated trajectory, Indicates the vehicle status when entering a curve.

8. The system according to claim 7, characterized in that Since the estimated target trajectory is used to generate feedback input, it is necessary to Convert to Cartesian coordinates , using a proportional-integral-derivative controller to track the trajectory Get additional input ; The end-to-end actions output by reinforcement learning will serve as important feedforward references and are recorded as ; Hierarchical decision-making has the same goal as data-driven decision-making, and uses direct addition to obtain the execution input: 。

Images