Unmanned vehicle control method and system based on minimum risk strategy
By classifying risk levels and formulating corresponding minimum risk strategies, the problem of strategy deficiency in L4 autonomous vehicles under fault conditions is solved, the reliability and safety of the system are improved, and the risk of rear-end collisions during emergency braking is reduced.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHERY AUTOMOBILE CO LTD
- Filing Date
- 2026-01-16
- Publication Date
- 2026-04-21
AI Technical Summary
Existing Level 4 autonomous vehicles lack hierarchical strategies and have vague operational details in their Minimum Risk Management (MRM) systems, resulting in insufficient strategy flexibility and a high risk of rear-end collisions during emergency braking.
By monitoring vehicle malfunction information, risk levels are classified, and different minimum risk strategies are formulated, including low risk, medium risk, high risk, and emergency risk. Vehicle routes are planned and corresponding operations are executed for each level. In case of emergency, V2X data is shared to avoid rear-end collisions, and passengers are alerted inside the vehicle.
It improves the reliability and stability of autonomous driving systems in complex environments, reduces the probability of rear-end collisions during emergency braking, and ensures the safety of vehicles and passengers.
Smart Images
Figure CN121893989A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of autonomous driving technology, and specifically to a method and system for controlling unmanned vehicles based on a minimum risk strategy. Background Technology
[0002] With the continuous development of autonomous driving technology, Level 4 autonomous Robotaxi has gradually become a research hotspot. Level 4 autonomous driving refers to a state where a vehicle can achieve driverless operation under specific conditions and environments, requiring no intervention from the driver, and the car can autonomously complete the journey from origin to destination. However, in actual operation, due to various factors such as failure of environmental perception functions, vehicle exceeding its designed operating area (ODD), and failure of the system planning control unit or actuators, the autonomous driving system may fail to perform dynamic driving tasks normally. In such cases, a minimum risk strategy (MRM) is required to ensure vehicle operation safety.
[0003] The relevant technologies in L4 autonomous driving mainly rely on multi-level redundancy design and real-time risk assessment. By fusing data from multiple sensors such as LiDAR, cameras, and millimeter-wave radar, it ensures that environmental risks can still be identified even if a single sensor fails. When the Design Operating Conditions (ODC) are about to be unmet, the Autonomous Driving System (ADS) promptly executes the Minimum Risk Principle (MRM) and enables the vehicle to come to a stop before the ODC is no longer met; if the ODC is suddenly unmet due to special circumstances, ADS immediately executes the Minimum Risk Principle (MRM) and enables the vehicle to come to a stop.
[0004] The current Minimum Risk Management (MRM) system design for autonomous driving still has several shortcomings: First, the MRM lacks a tiered strategy and does not clearly distinguish the specific measures corresponding to different road conditions or vehicle malfunctions, which may lead to insufficient strategy flexibility; second, the standard's definition of "reasonable control strategy" is rather vague and lacks specific operational details (such as deceleration rate and parking position requirements); and when encountering high-risk road scenarios and needing to brake suddenly, it may cause rear-end collisions, easily leading to collision hazards. Summary of the Invention
[0005] To address this, embodiments of the present invention provide a method and system for controlling unmanned vehicles based on a minimum risk strategy, in order to solve the problems of missing minimum risk strategy (MRM) grading strategies for unmanned Robotaxis, unclear corresponding specific operations, and the risk of rear-end collisions when executing emergency braking schemes.
[0006] To achieve the above objectives, the embodiments of the present invention provide the following technical solutions: According to a first aspect of the present invention, an embodiment of the present invention provides an autonomous vehicle control method based on a minimum risk strategy, the method comprising: The system monitors fault information that occurs during vehicle operation and classifies the corresponding risk levels based on the fault information. The risk levels include low risk, medium risk, high risk, and emergency risk, and different risk levels correspond to different minimum risk strategies. Control the vehicle's state to switch from normal driving state to minimum risk strategy execution state; If the risk level is low, medium, or high, a decision is made based on the minimum risk strategy corresponding to low, medium, or high risk, the vehicle's driving path is planned, and the vehicle is controlled to park according to the plan. If it is an emergency risk, the minimum risk strategy corresponding to the emergency risk will be executed. An emergency braking command will be sent directly to the CanBus module to control the vehicle to brake suddenly. After the vehicle brakes suddenly, data will be shared with surrounding vehicles via V2X to avoid rear-end collisions. Obtain the minimum strategy execution feedback result. If the feedback result indicates that the execution is complete, control the vehicle state to switch from the minimum risk strategy execution state to the minimum risk completion state.
[0007] Furthermore, different risk levels correspond to different minimum risk strategies, specifically including: If the risk level is low, the minimum risk strategy is to park in a safe area. If the risk level is medium, the least risky strategy is to pull over to the side of the road. If the risk is high, the least risky strategy is to slowly brake to a stop in the current lane; If it is an emergency risk, the least risky strategy is to brake to a stop in the current lane.
[0008] Furthermore, if the risk level is low, medium, or high, a decision is made based on the minimum risk strategy corresponding to that level, planning the vehicle's driving path and controlling the vehicle to park according to the plan. Specifically, this includes: If the risk level is low, the system will search for the nearest safe parking area within 1km on the high-precision map, with an estimated travel time of less than 3 minutes based on the road speed limit, and generate a route planning result. The system will then proceed to park in the safe area based on the route planning result. If no safe parking area meets the requirements, the system will report "No available safe parking area" and upgrade the risk level.
[0009] Furthermore, if the risk level is low, medium, or high, a decision is made based on the minimum risk strategy corresponding to that level, planning the vehicle's driving path and controlling the vehicle to park according to the plan. Specifically, this includes: If the risk level is medium, the vehicle will activate its hazard lights. When the nearest intersection on the navigation route is ≥ N*D+50 meters away from the vehicle, a pull-to-the-side parking task will be performed; where N is the number of lane changes and D is the longitudinal distance traveled in a single lane change. When the nearest intersection on the navigation route is < N*D+50 meters away from the vehicle, the vehicle will proceed normally along the ordered route through the intersection, and then perform the pull-to-the-side parking task. During the pull-to-the-side parking task, no-parking information from the high-precision map will be loaded to avoid no-parking areas. If the vehicle is unable to park at the original parking spot due to limitations in capability, road conditions, or traffic regulations during the pull-to-the-side parking process, a "failed to pull-to-the-side parking" message will be displayed, and the risk level will be upgraded.
[0010] Furthermore, if the risk level is low, medium, or high, a decision is made based on the minimum risk strategy corresponding to that level, planning the vehicle's driving path and controlling the vehicle to park according to the plan. Specifically, this includes: If the risk is high, a slow braking route will be generated, with the braking deceleration controlled within 3m / s², and the vehicle will be controlled to activate hazard lights. Priority will be given to avoiding congested traffic and fast-moving traffic lanes near intersections to ensure that there is no risk of collision during braking.
[0011] Furthermore, in the case of an emergency risk, the minimum risk strategy corresponding to the emergency risk is executed, directly sending an emergency braking command to the CanBus module to control the vehicle's emergency braking. After the vehicle brakes suddenly, data is shared with surrounding vehicles via V2X to prevent rear-end collisions. Specifically, this includes: A sudden braking command is sent directly to the CanBus module, controlling the vehicle to brake and come to a stop with a large deceleration along the current road direction. Specifically, when the vehicle speed is >60 kph, the deceleration magnitude is set to... When the vehicle speed is ≤60kph, set the deceleration magnitude to At the same time, a command to turn on the hazard lights is sent to the CanBus module to alert other vehicles and reduce the risk of collision.
[0012] Furthermore, the system obtains the minimum strategy execution feedback result. If the feedback indicates completion, the system switches the vehicle state from the minimum risk strategy execution state to the minimum risk completion state. This specifically includes: If the feedback result is incomplete or the execution timeout occurs when the risk level is low, medium or high, the risk level will be upgraded until the minimum risk strategy is completed. If the feedback result is that the execution is completed, the vehicle status will be switched to the minimum risk completed status, and the passengers in the vehicle will then confirm the vehicle status or request remote takeover. In case of emergency, once the vehicle has come to a complete stop, the CanBus module sends a chassis chassis signal, which includes a speed and stationary status indicator. If the chassis signal is detected and the execution result is incomplete, an emergency braking command is issued until the minimum risk strategy is completed. If the completion requirement is met, the vehicle status switches to the minimum risk completed state, and passengers inside the vehicle can then confirm the vehicle status or request remote takeover.
[0013] Furthermore, the system obtains the minimum strategy execution feedback result. If the feedback result indicates completion, the system switches the vehicle state from the minimum risk strategy execution state to the minimum risk completion state. Specifically, this also includes: When the feedback indicates completion, the vehicle status switches to the minimum risk completion state and performs the following actions: P-level control: Sends a P-level request to the CanBus module; Send a request to the CanBus module to engage the electronic parking brake (EPB); Hazard flasher control: Keep hazard flashers active; Reminder Trigger: Send a text and voice reminder to the HMI indicating that the execution is complete.
[0014] Furthermore, the method also includes: During the execution of the minimum risk strategy, the vehicle keeps its hazard lights on, and the in-vehicle voice prompts the current strategy until the vehicle comes to a complete stop. Passengers then decide whether to continue driving or receive remote assistance.
[0015] According to a second aspect of the present invention, an embodiment of the present invention provides an autonomous vehicle control system based on a minimum risk strategy, the system comprising: The fault diagnosis module is used to monitor fault information that occurs during vehicle operation, classify the corresponding risk level according to the fault information, including low risk, medium risk, high risk and emergency risk, and different risk levels correspond to different minimum risk strategies. The module generates minimum risk strategy event codes and sends them to the state machine module. The state machine module is used to execute vehicle state switching, send vehicle state information to associated modules, and receive feedback on the execution of the minimum risk strategy. The planning module is used to combine the path information provided by the routing module to generate corresponding obstacle avoidance or parking trajectories for different risk levels and send the planned trajectory information to the control module. The control module is used to convert the planned trajectory into action execution commands and send them to the CanBus module; The CanBus module is used to forward action execution commands to the vehicle chassis actuators; receive emergency braking commands from the state machine module and forward them to the actuators; and provide feedback on the execution results to the state machine module. The high-precision map module is used to provide high-precision map information; The routing module is used to generate path results based on high-precision map information and send them to the planning module; The human-computer interaction module is used to output corresponding text prompts and voice broadcasts based on different risk levels and strategies.
[0016] Compared with existing technologies, the autonomous vehicle control method and system based on the minimum risk strategy provided by this invention have the following beneficial effects: In the operation of autonomous robotaxis, by utilizing the fusion of data from multiple sensors such as LiDAR, cameras, and millimeter-wave radar, it ensures that environmental risks can still be identified even when a single sensor fails, improving the accuracy of identification in complex scenarios. When electronic components or software malfunctions occur in the vehicle, this invention classifies different risk levels based on the fault type and its impact on autonomous driving, and formulates multiple Minimum Risk Management (MRM) execution strategies. By defining multiple MRM strategies and linkage mechanisms with various modules, effective measures can be taken promptly when the autonomous driving system malfunctions or encounters abnormal situations, ensuring the safety of the vehicle and passengers. The system can automatically select appropriate MRM strategies based on different fault types and severity, and perform degradation processing, improving the reliability and stability of the autonomous driving system in complex environments. During the execution of MRM, the vehicle maintains hazard lights and shares data with surrounding vehicles via V2X, reducing the probability of rear-end collisions in emergency braking scenarios. Simultaneously, the vehicle provides voice and text reminders to passengers that it is currently implementing the minimum risk strategy, serving to alert and reassure them. Attached Figure Description
[0017] The accompanying drawings, which form part of this specification, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings: Figure 1 A schematic diagram illustrating the overall workflow of an autonomous vehicle control method based on a minimum risk strategy, provided in an embodiment of the present invention. Figure 2 A flowchart illustrating the interaction between modules of an autonomous vehicle control system based on a minimum risk strategy, provided for an embodiment of the present invention. Figure 3 This is a schematic diagram of the execution of a low-risk level MRM in an autonomous vehicle control method based on a minimum risk strategy provided in an embodiment of the present invention, hereinafter referred to as MRM_1, where the strategy is to park within a safe area; Figure 4 This is a schematic diagram of the medium-risk level MRM in an autonomous vehicle control method based on a minimum risk strategy provided in an embodiment of the present invention, hereinafter referred to as MRM_2, where the strategy is to pull over to the side of the road. Figure 5 This is a schematic diagram of the execution of a high-risk level MRM in an autonomous vehicle control method based on a minimum risk strategy provided in an embodiment of the present invention, hereinafter referred to as MRM_3, the strategy being to slowly brake to a stop in the current lane; Figure 6 This is a schematic diagram of the Emergency Risk Level (MRM) in an autonomous vehicle control method based on a minimum risk strategy provided in an embodiment of the present invention. Hereinafter referred to as MRM_4, the strategy is to brake to an emergency stop in the current lane. Detailed Implementation
[0018] The present invention will now be described in detail with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other.
[0019] The following detailed description is exemplary and intended to provide further detailed explanation of the invention. Unless otherwise specified, all technical terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used in this invention is for describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention.
[0020] The first embodiment of this invention provides a method for controlling autonomous vehicles based on a minimum risk strategy. The following is a combination of... Figure 1 Please provide a detailed explanation.
[0021] like Figure 1 As shown, in step S100, fault information occurring during vehicle operation is monitored, and the corresponding risk level is divided according to the fault information. The risk level includes low risk, medium risk, high risk, and emergency risk, and different risk levels correspond to different minimum risk strategies.
[0022] like Figure 1 As shown, in step S200, the vehicle state is switched from normal driving state to minimum risk strategy execution state.
[0023] like Figure 1 As shown, in step S300, if the risk is low, medium or high, a decision is made based on the minimum risk strategy corresponding to low, medium or high risk, the vehicle travel path is planned, and the vehicle is controlled to park according to the plan.
[0024] like Figure 1 As shown, in step S400, if it is an emergency risk, the minimum risk strategy corresponding to the emergency risk is executed, and an emergency braking command is directly sent to the CanBus module to control the vehicle to brake suddenly. After the vehicle brakes suddenly, data is shared with surrounding vehicles via V2X to avoid rear-end collisions.
[0025] like Figure 1 As shown, in step S500, the minimum strategy execution feedback result is obtained. If the feedback result indicates that the execution is completed, the vehicle state is controlled to switch from the minimum risk strategy execution state to the minimum risk completion state.
[0026] Furthermore, the method also includes: during the execution of the minimum risk strategy, the vehicle keeps its hazard lights on, and the in-vehicle voice announces the current execution strategy until the vehicle comes to a complete stop, at which point the passengers in the vehicle decide whether to continue driving or provide remote driving assistance.
[0027] The specific configuration of the autonomous vehicle control system based on the minimum risk strategy in this invention embodiment is as follows: Figure 2 As shown, the method and process are explained below in conjunction with each module of the system, as detailed below: Step S1: The diagnostic module collects and sends the execution strategy type corresponding to the fault to the state machine module.
[0028] During autonomous driving, the diagnostic module receives fault information from upstream modules through communication with other modules and hardware devices, converts this fault information into DTCs (Diagnostic Trouble Codes), and supports the persistence of DTCs. When the diagnostic module receives this fault information, it analyzes it. If the fault affects the vehicle's autonomous driving system capabilities, the fault information is converted into a condition that triggers the MRM (Most Automated Management Response) and forwarded to the state machine module in the form of an MRM event code.
[0029] The state machine module receives the execution strategy type result from the diagnostic module. It manages MRM state transitions, such as switching from normal driving to executing the minimum risk strategy (MRM) and then to the completed state (MRC). Its function is to ensure that state transitions conform to safety logic. It also sends the current vehicle state to other related modules and receives MRM execution feedback results from the planning module and the CanBus module. When the MRM execution feedback result is successful, the state machine switches to the minimum risk completed state (MRC) and performs the following actions: 1. P-gear control: Sends a P-gear request to the CanBus module. 2. Send an electronic parking brake (EPB) activation request to the CanBus module. 3. Hazard flasher control: Keep hazard flashers active. 4. Reminder Trigger: Send a text and voice reminder indicating the completion status of MRM_4 execution to the Human-Computer Interaction (HMI) module. Step S2: The state machine module sends the current vehicle status information to the planning module.
[0030] The planning module generates specific obstacle avoidance or parking trajectories based on the received vehicle status information, satisfying dynamic feasibility and prioritizing safety over comfort when implementing the minimum risk strategy (MRM) mode.
[0031] In this design, the planning module execution strategies differ for different risk levels of MRM: Low-level minimum risk strategy (MRM_1), such as Figure 3 As shown: The routing module searches the high-precision map (step S4) for the nearest safe parking area within 1km and 3 minutes (estimated based on road speed limits), and generates a path result which is then sent to the planning module (step S5). The planning module executes a parking strategy towards the safe area based on this result. If no safe parking area meets the requirements, it sends a "no available safe parking area" message to the state machine module, which then performs a safety degradation state transition, for example, shifting from a low-risk minimum risk strategy to a medium-risk level. Medium-level minimum risk strategy (MRM_2), such as Figure 4 As shown: When the planning module receives the MRM_2 state from the state machine, the vehicle activates its hazard lights. If the nearest intersection on the navigation route is ≥ N*D + 50m away from the vehicle, it executes a roadside parking task; where N is the number of lane changes and D is the longitudinal distance of a single lane change (the distance varies depending on the speed). If the nearest intersection on the navigation route is < N*D + 50m away from the vehicle, the vehicle proceeds normally through the intersection according to the ordered route, and then executes the roadside parking task. During the roadside parking process, the planning module automatically loads no-parking information from the yellow grid lines on the high-precision map from the disk and plans a route to avoid no-parking areas based on this information. If, during the roadside parking process, the vehicle cannot stop at the original parking spot due to capacity / road conditions / traffic regulations, it sends a "failed to park" message to the state machine module, which then performs a safety degradation state transition.
[0032] High-level minimum risk strategy (MRM_3), such as Figure 5 As shown: When the planning module receives the MRM_3 state from the state machine module, it will select a slow braking route on the existing planned trajectory. The braking deceleration should be controlled within... Within this range, the system will activate hazard lights to ensure a comfortable stop within the lane. Ensure there is no risk of collision during braking. Prioritize parking locations away from intersections, congested traffic lanes, and fast-moving traffic lanes.
[0033] Step S3: The state machine module sends the current vehicle state information to the human-machine interaction module. The Human-Machine Interface (HMI) module receives vehicle status information from the state machine module and provides different text prompts and voice announcements based on different risk levels, for example: When the low-level minimum risk strategy (MRM_1) is executed: HMI [Text Display] "Proceeding to a safe parking area" HMI [Voice Prompt] "Proceeding to a safe parking area" (This is said once when the function is triggered; it does not need to be repeated during function execution). Warning flashing; the screen flashes red as a warning (lasts up to 3 seconds). When the low-level minimum risk strategy (MRM_1) is completed: HMI [Classical Chinese] "The vehicle has come to a stop. Please immediately check the vehicle's status or request remote takeover." HMI [Voice Prompt] "The vehicle has come to a stop. Please check the vehicle status immediately or request remote takeover." Step S4: The high-precision map module sends the location information that meets the criteria to the routing module: Based on the requirements of the planning module, the location information that matches the current risk level is found on the high-precision map and sent to the routing module. For example, executing MRM1 requires finding the nearest safe parking area within 1km and 3 minutes away. Executing MRM2 requires a suitable roadside parking location. Executing MRM3 requires a location on the current path where the vehicle can slow down and stop.
[0034] Step S5: The routing module generates a path result based on the destination location and the high-precision map, and then provides it to the planning module.
[0035] Step S6: The planning module sends the planning trajectory information of the MRM execution to the control module.
[0036] Step S7: The control module converts the planned trajectory into execution commands (steering angle, braking pressure), sends them to the Canbus module, and then the Canbus communicates with the chassis and sends them to achieve precise vehicle control.
[0037] Step S8: The planning module feeds back the execution results of low, medium, and high risk levels of MRM to the state machine module: When the planning module executes the Minimum Risk Completion (MRM), it sends the execution result back to the state machine. If the feedback result is incomplete or the execution times out, the state machine performs a safety risk downgrade until the MRM execution is complete. If the feedback result indicates completion, the state machine switches to the Minimum Risk Completion (MRC) state, and then passengers in the vehicle confirm the vehicle status or request remote takeover. At this point, the minimum risk strategy MRM process is complete. The conditions for MRM execution to complete are explained below: 1. The parking result meets the parking entry and exit standards: 2. Park your vehicle in a safe area / the rightmost lane (if the rightmost lane is full, temporarily park in the next rightmost lane). 3. Vehicles should avoid no-parking areas (within intersections and yellow no-parking grid areas). 4. The vehicle is fully inside the rightmost lane. 5. After the vehicle has come to a complete stop, keep the hazard lights on and ensure the vehicle does not roll back within 3 seconds of coming to a complete stop. 6. The vehicle body should be aligned with the road direction, and the direction in which the front of the vehicle is facing should not deviate from the center line of the road by more than 5 degrees. Step S9: The state machine module sends control commands to the CanBus module: like Figure 6 As shown, when the vehicle is in Emergency Risk Level MRM_4, the status machine will send an emergency braking command directly to the CanBus module, braking and coming to a stop with a large deceleration along the current road direction. When the vehicle speed is >60kph, the deceleration magnitude can be set to... When the vehicle speed is ≤60kph, the deceleration magnitude can be set to... The state machine sends a command to the CanBus module to activate the hazard warning lights, alerting other vehicles to reduce the risk of collision.
[0038] Step S10: The CanBus module feeds back the execution result of the high-risk level MRM_4 to the state machine module: The condition for a vehicle to complete Emergency Risk Level MRM_4 is that the vehicle comes to a complete stop and remains stationary for more than 3 seconds. After emergency braking, the vehicle shares data with surrounding vehicles via V2X to prevent rear-end collisions. Once the vehicle has come to a complete stop, the CanBus module sends a chassis chassis signal to the state machine, which includes a speed and stationary status flag. The state machine checks the chassis signal; if the feedback indicates incomplete completion, it continues to issue emergency braking commands until MRM_4 is completed. If the completion requirements are met, the state machine switches to the Minimum Risk Completion (MRC) state, and passengers inside the vehicle confirm the vehicle's status or request remote takeover, thus completing the Minimum Risk Strategy (MRM) process.
[0039] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.
Claims
1. A method for controlling an autonomous vehicle based on a minimum risk strategy, characterized in that, The method includes: The system monitors fault information that occurs during vehicle operation and classifies the corresponding risk levels based on the fault information. The risk levels include low risk, medium risk, high risk, and emergency risk, and different risk levels correspond to different minimum risk strategies. Control the vehicle's state to switch from normal driving state to minimum risk strategy execution state; If the risk level is low, medium, or high, a decision is made based on the minimum risk strategy corresponding to low, medium, or high risk, the vehicle's driving path is planned, and the vehicle is controlled to park according to the plan. If it is an emergency risk, the minimum risk strategy corresponding to the emergency risk will be executed, and an emergency braking command will be sent directly to the CanBus module to control the vehicle to brake suddenly. After the vehicle brakes suddenly, data will be shared with surrounding vehicles via V2X to avoid rear-end collisions. Obtain the minimum strategy execution feedback result. If the feedback result indicates that the execution is complete, control the vehicle state to switch from the minimum risk strategy execution state to the minimum risk completion state.
2. The autonomous vehicle control method based on a minimum risk strategy according to claim 1, characterized in that, Different risk levels correspond to different minimum risk strategies, specifically including: If the risk level is low, the minimum risk strategy is to park in a safe area. If the risk level is medium, the least risky strategy is to pull over to the side of the road. If the risk is high, the least risky strategy is to slowly brake to a stop in the current lane; If it is an emergency risk, the least risky strategy is to brake to a stop in the current lane.
3. The autonomous vehicle control method based on a minimum risk strategy according to claim 2, characterized in that, If the risk level is low, medium, or high, a decision is made based on the minimum risk strategy corresponding to that level, including planning the vehicle's driving path and controlling the vehicle to park according to the plan. This specifically includes: If the risk level is low, the system will search for the nearest safe parking area within 1km on the high-precision map, with an estimated travel time of less than 3 minutes based on the road speed limit, and generate a route planning result. The system will then proceed to park in the safe area based on the route planning result. If no safe parking area meets the requirements, the system will report "No safe parking area available" and upgrade the risk level.
4. The autonomous vehicle control method based on a minimum risk strategy according to claim 2, characterized in that, If the risk level is low, medium, or high, a decision is made based on the minimum risk strategy corresponding to that level, including planning the vehicle's driving path and controlling the vehicle to park according to the plan. This specifically includes: If the risk level is medium, the vehicle will activate its hazard lights. When the nearest intersection on the navigation route is ≥ N*D+50 meters away from the vehicle, a roadside parking task will be performed; where N is the number of lane changes and D is the longitudinal distance traveled in a single lane change. When the nearest intersection on the navigation route is < N*D+50 meters away from the vehicle, the vehicle will proceed normally along the ordered route through the intersection, and then perform the roadside parking task. During the roadside parking task, no-parking information from the high-precision map will be loaded to avoid no-parking areas. If the vehicle is unable to park at the original parking spot due to limitations in capability, road conditions, or traffic regulations, a "failed to park" message will be displayed, and the risk level will be upgraded.
5. The autonomous vehicle control method based on a minimum risk strategy according to claim 2, characterized in that, If the risk level is low, medium, or high, a decision is made based on the minimum risk strategy corresponding to that level, including planning the vehicle's driving path and controlling the vehicle to park according to the plan. This specifically includes: If the risk is high, a slow braking route will be generated, with the braking deceleration controlled within 3m / s², and the vehicle will be controlled to activate hazard lights. Priority will be given to avoiding congested traffic and fast-moving traffic lanes near intersections to ensure that there is no risk of collision during braking.
6. The autonomous vehicle control method based on a minimum risk strategy according to claim 2, characterized in that, If it is an emergency risk, the minimum risk strategy corresponding to the emergency risk is executed, directly sending an emergency braking command to the CanBus module to control the vehicle to brake suddenly. After the vehicle brakes suddenly, data is shared with surrounding vehicles via V2X to prevent rear-end collisions. Specifically, this includes: A sudden braking command is sent directly to the CanBus module, controlling the vehicle to brake and come to a stop with a large deceleration along the current road direction. Specifically, when the vehicle speed is >60 kph, the deceleration magnitude is set to... When the vehicle speed is ≤60kph, set the deceleration magnitude to At the same time, a command to turn on the hazard lights is sent to the CanBus module to alert other vehicles and reduce the risk of collision.
7. The autonomous vehicle control method based on a minimum risk strategy according to claim 1, characterized in that, Obtain the minimum risk strategy execution feedback result. If the feedback result indicates completion, switch the vehicle state from the minimum risk strategy execution state to the minimum risk completion state. Specifically, this includes: If the feedback result is incomplete or the execution timeout occurs when the risk level is low, medium or high, the risk level will be upgraded until the minimum risk strategy is completed. If the feedback result is that the execution is completed, the vehicle status will be switched to the minimum risk completed status, and the passengers in the vehicle will then confirm the vehicle status or request remote takeover. In case of emergency, once the vehicle has come to a complete stop, the CanBus module sends a chassis chassis signal, which includes a speed and stationary status indicator. If the chassis signal is detected and the execution result is incomplete, an emergency braking command is issued until the minimum risk strategy is completed. If the completion requirement is met, the vehicle status switches to the minimum risk completed state, and passengers inside the vehicle can then confirm the vehicle status or request remote takeover.
8. The autonomous vehicle control method based on a minimum risk strategy according to claim 1, characterized in that, Obtain the minimum risk strategy execution feedback result. If the feedback result indicates completion, switch the vehicle state from the minimum risk strategy execution state to the minimum risk completion state. Specifically, this also includes: When the feedback indicates completion, the vehicle status switches to the minimum risk completion state and performs the following actions: P-level control: Sends a P-level request to the CanBus module; Send a request to the CanBus module to engage the electronic parking brake (EPB); Hazard flasher control: Keep hazard flashers active; Reminder Trigger: Send a text and voice reminder to the HMI indicating that the execution is complete.
9. The autonomous vehicle control method based on a minimum risk strategy according to claim 1, characterized in that, The method further includes: During the execution of the minimum risk strategy, the vehicle keeps its hazard lights on, and the in-vehicle voice prompts the current strategy until the vehicle comes to a complete stop. Passengers then decide whether to continue driving or receive remote assistance.
10. An autonomous vehicle control system based on a minimum risk strategy, characterized in that, The system includes: The fault diagnosis module is used to monitor fault information that occurs during vehicle operation, classify the corresponding risk level according to the fault information, including low risk, medium risk, high risk and emergency risk, and different risk levels correspond to different minimum risk strategies. The module generates minimum risk strategy event codes and sends them to the state machine module. The state machine module is used to execute vehicle state switching, send vehicle state information to associated modules, and receive feedback on the execution of the minimum risk strategy. The planning module is used to combine the path information provided by the routing module to generate corresponding obstacle avoidance or parking trajectories for different risk levels and send the planned trajectory information to the control module. The control module is used to convert the planned trajectory into action execution commands and send them to the CanBus module; The CanBus module is used to forward action execution commands to the vehicle chassis actuators; receive emergency braking commands from the state machine module and forward them to the actuators; and provide feedback on the execution results to the state machine module. The high-precision map module is used to provide high-precision map information; The routing module is used to generate path results based on high-precision map information and send them to the planning module; The human-computer interaction module is used to output corresponding text prompts and voice broadcasts based on different risk levels and strategies.