An Automatic Rescue Method for Low-Speed Autonomous Vehicles

Abstract

The present invention relates to an automatic rescue method for low-speed autonomous vehicles, which includes the following steps: after a failure occurs in an autonomous vehicle, the failure level of the faulty vehicle and the detailed information of the faulty vehicle are reported to the cloud-based autonomous driving management system; the cloud-based autonomous driving management system determines a rescue plan according to the failure situation of the faulty vehicle. If the failure of the faulty vehicle belongs to a type that cannot be automatically rescued, manual rescue is notified; if the failure of the faulty vehicle belongs to a type that can be automatically rescued, automatic rescue is performed on the faulty vehicle; after determining to implement automatic rescue, a rescue vehicle is designated to go to perform the rescue task; after the rescue vehicle travels to the location of the faulty vehicle, it establishes a vehicle-to-vehicle wireless connection with the faulty vehicle through identity verification and enters the wireless traction mode; the faulty vehicle is towed to the target station or repair shop, and the wireless traction mode is released. The automatic rescue method of the present invention does not require manual participation and can improve the rescue efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of autonomous driving, and particularly to an automatic rescue method for low-speed autonomous driving vehicles. Background Art

[0002] In recent years, the autonomous driving technology has developed rapidly. The most representative technology is the single-vehicle intelligence technology. Generally speaking, various types of sensor devices are installed on an autonomous driving vehicle, and the on-vehicle computing unit is used to sense the surrounding obstacles and road conditions. At the same time, through the on-vehicle computing unit, a feasible decision-making and planning result is calculated to achieve the purpose of autonomous driving. In addition, the autonomous driving industry has also begun to try to build fixed v2x device modules on demonstration roads to make up for the deficiencies of the on-vehicle hardware capabilities of single vehicles.

[0003] During the driving process of autonomous driving vehicles, faults often occur. The faulty vehicles may not be able to drive autonomously to the repair station for repair, so it is necessary to rescue the faulty vehicles.

[0004] Currently, in the case of a chassis fault of an autonomous driving vehicle, basically only manual means can be used to transport the vehicle back to the repair shop. However, in the case of no chassis fault, there are currently only three means that can be used. One is that the operator remotely controls the vehicle to return to the station or the repair shop on-site. The second is to remotely control the vehicle to return to the station or the repair shop through the method of remote video and remote control instructions. The last one is to directly call the towing service to transport the vehicle back to the station or the repair shop. The disadvantages of the existing rescue methods are that they all require manual intervention to assist in returning the faulty vehicle to the station or the repair shop, and there is a relatively high labor cost. Summary of the Invention

[0005] The purpose of the present invention is to provide an automatic rescue method for low-speed autonomous driving vehicles in view of the problem in the prior art described in the background art that when an autonomous driving single vehicle fails and needs to be rescued, manual intervention is required and the labor cost is high.

[0006] To achieve the above purpose, the present invention is realized through the following technical solutions: An automatic rescue method for low-speed autonomous driving vehicles, which includes the following steps:

[0007] S1. After a fault occurs in an autonomous driving vehicle, the fault level of the faulty vehicle and the detailed information of the faulty vehicle are reported to the cloud autonomous driving management system;

[0008] S2. The cloud autonomous driving management system determines a rescue plan according to the fault situation of the faulty vehicle. If the fault of the faulty vehicle belongs to a type that cannot be automatically rescued, it notifies manual rescue. If the fault of the faulty vehicle belongs to a type that can be automatically rescued, it conducts automatic rescue on the faulty vehicle;

[0009] S3. After the automatic rescue is determined, the cloud-based automatic driving management system dispatches nearby automatic driving vehicles according to the location of the faulty vehicle, and designates a rescue vehicle to perform the rescue mission according to the actual situation of the automatic driving vehicle. The goal of the mission is to rescue the faulty vehicle back to the station or maintenance point;

[0010] S4. After the rescue vehicle reaches the location of the faulty vehicle, it establishes a wireless vehicle-to-vehicle connection with the faulty vehicle through identity authentication, enters the wireless traction mode, and requisitions the sensors and computing resources available to the faulty vehicle. By means of capacity contribution, it integrates all available resources. During the rescue mission, the rescue vehicle calculates the rescue vehicle's driving trajectory and the faulty vehicle's following trajectory and control instructions, and controls the two vehicles to complete the automatic rescue.

[0011] S5. After the rescue vehicle wirelessly tows the faulty vehicle to the target site or repair shop, the wireless towing mode will be automatically released, and the completion of the rescue mission will be reported, and then the autonomous driving mission will continue.

[0012] As a further improvement to the above solution, the autonomous driving vehicle is provided with a diagnostic module for analyzing vehicle faults. The fault information diagnosed by the diagnostic module includes vehicle information, time and space information, fault type and irrecoverable cause. By setting up the diagnostic module, when an autonomous driving vehicle fails, the fault of the autonomous driving vehicle and the cause of the failure can be automatically diagnosed. If the fault cannot be recovered by the autonomous driving vehicle itself, it needs to be reported to the cloud-based autonomous driving management system.

[0013] As a further improvement of the above scheme, if the communication between the faulty vehicle and the cloud-based autonomous driving management system is normal, the faulty vehicle will actively report the fault to the cloud-based autonomous driving management system after the fault occurs; if there is no network connection or the connection fails when the faulty vehicle fails, it will wait for other passing autonomous driving vehicles, contact the passing vehicles through the emergency vehicle-to-vehicle communication channel, and report the fault to the cloud-based autonomous driving management platform through the network of other passing autonomous driving vehicles. Through this setting, the fault information of the faulty vehicle can be reported to the cloud-based autonomous driving management system in a timely manner, improving the rescue efficiency of the faulty vehicle.

[0014] As a further improvement of the above solution, the cloud-based autonomous driving management system can determine whether a vehicle is out of contact based on the heartbeat data of each vehicle. After determining that a vehicle is out of contact, the cloud-based autonomous driving management platform will actively dispatch the vehicles around the last known location of the out-of-contact vehicle to search for the faulty vehicle along the mission route of the out-of-contact vehicle, so as to detect the faulty vehicle in a timely manner. After detecting the faulty vehicle, a vehicle-to-vehicle communication is established between the vehicle that detected the faulty vehicle and the faulty vehicle, and the fault information of the faulty vehicle is uploaded to the cloud-based autonomous driving management system. If the fault type belongs to the type that cannot be automatically rescued, manual rescue will be notified; if the fault of the faulty vehicle belongs to the type that can be automatically rescued, the cloud-based autonomous driving management system will assign an automatic rescue task to the vehicle that detected the faulty vehicle to automatically rescue the faulty vehicle. Through this setting, the faulty vehicle can be detected faster and automatically rescued.

[0015] As a further improvement of the above solution, in step S4, the available sensors of the faulty vehicle include one or more of lidar, camera, GPS, IMU, ultrasonic radar, millimeter-wave radar, and anti-collision strips.

[0016] As a further improvement of the above solution, in step S4, the computing power resources include one or more of CPU, GPU, and FPGA.

[0017] As a further improvement of the above solution, in step S4, if the sensors and computing power resources of the faulty vehicle can work properly and can be shared and used by the rescue vehicle, the rescue vehicle will uniformly dispatch and use the sensors and computing power resources of the faulty vehicle to assist in the autonomous driving calculation of the wireless traction mode; if the sensors and computing power resources of the faulty vehicle cannot be shared and used by the rescue vehicle, the rescue vehicle can only use the sensors and computing power resources of its own vehicle to complete the rescue of the faulty vehicle.

[0018] The present invention has positive effects: The automatic rescue method for low-speed autonomous driving vehicles of the present invention can perform rescue according to the fault type of the faulty vehicle. If the faulty vehicle belongs to the type that cannot be automatically rescued, manual rescue will be arranged; if it belongs to the type that can be automatically rescued, a rescue vehicle will be assigned. Through vehicle-to-vehicle communication, the sensors and computing power resources of the faulty vehicle and the sensors and computing power resources of the rescue vehicle are integrated, and the rescue vehicle wirelessly towes the faulty vehicle to the target station or repair station. Through the improvement of the present invention, case-by-case automatic rescue of the faulty vehicle can be realized, reducing manual consumption and improving rescue efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is the working principle diagram of the automatic rescue method for low-speed autonomous driving vehicles of the present invention. The numbers in the figure are the implementation sequence of the automatic rescue method for low-speed autonomous driving vehicles of the present invention. Detailed implementation manners

[0020] The technical solutions of the present invention will be clearly and completely described below through embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0021] An automatic rescue method for a low-speed autonomous vehicle, which includes the following steps:

[0022] S1. After a failure occurs in the autonomous vehicle, the failure level of the faulty vehicle and the detailed information of the faulty vehicle are reported to the cloud autonomous driving management system.

[0023] Diagnostic modules for fault analysis of the vehicle are provided on all autonomous vehicles. The fault information diagnosed by the diagnostic modules includes vehicle information, spatio-temporal information, fault types, and reasons for non-recovery. By setting the diagnostic module, when a failure occurs in the autonomous vehicle, the fault of the autonomous vehicle and the cause of the failure can be automatically diagnosed. If the fault cannot be recovered by the autonomous vehicle itself, it is necessary to report to the cloud autonomous driving management system.

[0024] According to the fault situation of the faulty vehicle, if normal communication can be established between the faulty vehicle and the cloud autonomous driving management system, the diagnosis result of the diagnostic module of the faulty vehicle can be directly uploaded to the cloud autonomous driving management system; if a communication failure occurs between the faulty vehicle and the cloud autonomous driving management system, there is no network connection or the connection fails, then the fault of the faulty vehicle cannot be directly uploaded to the cloud autonomous driving management system, and it is necessary to wait for other passing autonomous vehicles, contact the passing vehicles through the emergency vehicle-to-vehicle communication channel, and report the fault to the cloud autonomous driving management platform through the network of other passing autonomous vehicles. Through this setting, the fault information of the faulty vehicle can be reported to the cloud autonomous driving management system in a timely manner, improving the rescue efficiency of the faulty vehicle.

[0025] In addition to passively waiting for other autonomous vehicles passing by, the cloud-based autonomous driving management system can determine whether a vehicle is out of contact based on the heartbeat data of each vehicle. After determining that a vehicle is out of contact, the cloud-based autonomous driving management platform will actively dispatch the vehicles around the last known location of the out-of-contact vehicle to search for the faulty vehicle along the task route of the out-of-contact vehicle, so as to detect the faulty vehicle in a timely manner. After detecting the faulty vehicle, a vehicle-to-vehicle communication is established between the vehicle that detected the faulty vehicle and the faulty vehicle, and the fault information of the faulty vehicle is uploaded to the cloud-based autonomous driving management system. If the fault type belongs to the type that cannot be automatically rescued, manual rescue will be notified; if the fault of the faulty vehicle belongs to the type that can be automatically rescued, the cloud-based autonomous driving management system will assign an automatic rescue task to the vehicle that detected the faulty vehicle to automatically rescue the faulty vehicle. Through this setting, faulty vehicles can be detected faster and automatically rescued.

[0026] S2. The cloud-based autonomous driving management system determines the rescue plan according to the fault situation of the faulty vehicle. If the fault of the faulty vehicle belongs to the type that cannot be automatically rescued, manual rescue will be notified. The types of faults that cannot be automatically rescued mentioned here include situations where the autonomous vehicle cannot drive, such as the situation where the chassis of the autonomous vehicle fails and cannot drive.

[0027] If the fault of the faulty vehicle belongs to the type that can be automatically rescued, the faulty vehicle will be automatically rescued; the types of faults that can be automatically rescued usually refer to the situations where the chassis and vehicle-to-vehicle communication module of the autonomous vehicle can work normally, and other systems have irrecoverable faults, and these situations are not of the types that can be automatically rescued.

[0028] S3. After determining to implement automatic rescue, the cloud-based autonomous driving management system schedules the nearby autonomous vehicles according to the location of the faulty vehicle, and designates a rescue vehicle to go to perform the rescue task according to the actual situation of the autonomous vehicle. The goal of the task is to rescue the faulty vehicle back to the station or repair point.

[0029] After the rescue vehicle travels to the location of the faulty vehicle, it establishes a vehicle-to-vehicle wireless connection with the faulty vehicle through identity verification, enters the wireless traction mode, and commandeers the available sensors and computing power resources of the faulty vehicle. By means of capacity contribution, all available resources are integrated. During the process of the rescue vehicle performing the rescue task, it will calculate the driving of the rescue vehicle and the following trajectory and control instructions of the faulty vehicle to control the two vehicles to complete the automatic rescue. The above-mentioned identity verification method can adopt the method of verifying the identity key.

[0030] The available sensors of the faulty vehicle include one or several of lidar, camera, gps, imu, ultrasonic radar, millimeter wave radar, and anti-collision strips. The computing power resources include one or several of cpu, gpu, and fpga.

[0031] If the sensors and computing resources of the faulty vehicle are able to work normally and can be shared by rescue vehicles, the rescue vehicles will uniformly dispatch and use the sensors and computing resources of the faulty vehicle to assist in the autonomous driving calculations of the wireless traction mode; if the sensors and computing resources of the faulty vehicle cannot be shared by rescue vehicles, the rescue vehicle can only use its own sensors and computing resources to complete the rescue of the faulty vehicle.

[0032] S5. After the rescue vehicle wirelessly tows the faulty vehicle to the target site or repair shop, the wireless towing mode will be automatically released, and the completion of the rescue mission will be reported, and then the autonomous driving mission will continue.

[0033] The automatic rescue method for low-speed automatic driving vehicles of the present invention performs unified dispatch management of automatic driving vehicles through a cloud-based automatic driving management system. When an automatic driving vehicle fails, if the faulty vehicle can communicate with the cloud-based automatic driving management system, the fault information obtained by the diagnosis module can be uploaded to the cloud-based automatic driving management system. According to the fault type, if the fault of the faulty vehicle is a type that cannot be automatically rescued, the cloud-based automatic driving management system notifies manual rescue of the faulty vehicle. If the fault of the faulty vehicle is a type that can be automatically rescued, the cloud-based automatic driving management system allocates rescue vehicles according to priority. After the rescue vehicle arrives near the faulty vehicle, it can establish vehicle-to-vehicle communication with the faulty vehicle, and tow the faulty vehicle to the target site or maintenance station for repair by wireless traction. Through the improvement of the present invention, after an automatic driving vehicle fails, it can be classified and rescued according to the fault type, and the rescue efficiency is higher. Moreover, for the type that can be automatically rescued, no human participation is required, which reduces labor consumption and improves rescue efficiency.

[0034] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An automatic rescue method for low-speed autonomous vehicles, characterized in that: It includes the following steps: S1. After an autonomous driving vehicle fails, the fault level and detailed information of the faulty vehicle are reported to the cloud-based autonomous driving management system; S2. The cloud-based autonomous driving management system determines a rescue plan based on the fault condition of the faulty vehicle. If the fault of the faulty vehicle is of a type that cannot be automatically rescued, manual rescue is notified; if the fault of the faulty vehicle is of a type that can be automatically rescued, the faulty vehicle is automatically rescued; S3. After the automatic rescue is determined, the cloud-based automatic driving management system dispatches nearby automatic driving vehicles according to the location of the faulty vehicle, and designates a rescue vehicle to perform the rescue mission according to the actual situation of the automatic driving vehicle. The goal of the mission is to rescue the faulty vehicle back to the station or maintenance point; S4. After the rescue vehicle reaches the location of the faulty vehicle, it establishes a wireless vehicle-to-vehicle connection with the faulty vehicle through identity authentication, enters the wireless traction mode, and requisitions the sensors and computing resources available to the faulty vehicle. By means of capacity contribution, it integrates all available resources. During the rescue mission, the rescue vehicle calculates the rescue vehicle's driving trajectory and the faulty vehicle's following trajectory and control instructions, and controls the two vehicles to complete the automatic rescue. S5. After the rescue vehicle wirelessly tows the faulty vehicle to the target site or repair shop, it will automatically cancel the wireless towing mode, report the completion of the rescue mission, and then continue to perform the autonomous driving mission; If the communication between the faulty vehicle and the cloud-based autonomous driving management system is normal, the faulty vehicle will actively report the fault to the cloud-based autonomous driving management system after the fault occurs; if there is no network connection or the connection fails when the faulty vehicle fails, it will wait for other passing autonomous driving vehicles, contact the passing vehicles through the emergency vehicle-to-vehicle communication channel, and report the fault to the cloud-based autonomous driving management platform through the network of other passing autonomous driving vehicles; In step S4, if the sensors and computing resources of the faulty vehicle can work normally and can be shared by the rescue vehicles, the rescue vehicles will uniformly dispatch and use the sensors and computing resources of the faulty vehicle to assist in the automatic driving calculation of the wireless traction mode; if the sensors and computing resources of the faulty vehicle cannot be shared by the rescue vehicles, the rescue vehicles can only use their own sensors and computing resources to complete the rescue of the faulty vehicle.

2. The automatic rescue method for low-speed autonomous driving vehicles according to claim 1, characterized in that: The autonomous driving vehicles are all provided with a diagnostic module for performing fault analysis on the vehicle. The fault information diagnosed by the diagnostic module includes vehicle information, time and space information, fault type and irrecoverable cause.

3. The automatic rescue method for low-speed autonomous vehicles according to claim 1, characterized in that: The cloud-based autonomous driving management system can determine whether a vehicle is out of contact based on the heartbeat data of each vehicle. After determining that a vehicle is out of contact, the cloud-based autonomous driving management platform will actively dispatch the vehicles around the last known position of the out-of-contact vehicle to search for the faulty vehicle along the task route of the out-of-contact vehicle, so as to detect the faulty vehicle in a timely manner. After detecting the faulty vehicle, vehicle-to-vehicle communication is established between the vehicle that detected the faulty vehicle and the faulty vehicle, and the fault information of the faulty vehicle is uploaded to the cloud-based autonomous driving management system. If the fault type belongs to the type that cannot be automatically rescued, manual rescue is notified; if the fault of the faulty vehicle belongs to the type that can be automatically rescued, the cloud-based autonomous driving management system assigns an automatic rescue task to the vehicle that detected the faulty vehicle to automatically rescue the faulty vehicle.

4. The automatic rescue method for a low-speed autonomous vehicle according to claim 1, characterized in that: In step S4, the available sensors of the faulty vehicle include one or more of lidar, camera, GPS, IMU, ultrasonic radar, millimeter wave radar, and anti-collision strips.

5. The automatic rescue method for a low-speed autonomous vehicle according to claim 1, characterized in that: In step S4, the computing power resources include one or more of CPU, GPU, and FPGA.

Images