Cooperative unmanned aerial vehicle accompanying flight system and method for rail transit pushing operation

Abstract

The invention provides a cooperative unmanned aerial vehicle accompanying flight system and method for rail transit pushing operation. The system comprises an unmanned aerial vehicle, a cooperative control subsystem, a ground command subsystem and an operation execution subsystem. The unmanned aerial vehicle subsystem is used for keeping a relative position with the train for accompanying flight in the pushing operation, acquiring images and distance measurement data of the train and a line environment, acquiring measurement data of a vehicle connecting part, and realizing fixation or separation with a train body through a gripping apparatus structure; the ground command subsystem is connected with the cooperative control subsystem, and is used for generating a path plan of the push operation, and adjusting the path plan according to the obstacle and path deviation information in the safety monitoring result to generate a path adjustment instruction; and the operation execution subsystem is connected with the cooperative control subsystem and is used for moving to a target position according to the position information of the vehicle connection component to execute mechanical connection or disconnection operation between the vehicles. A closed loop of real-time information interaction, dynamic collaborative decision and automatic precise execution in the pushing operation process can be realized.

Claims

1. A cooperative unmanned aerial vehicle (UAV) escort system for rail transit pushing operations, characterized in that, include: The unmanned aerial vehicle (UAV) subsystem is used to maintain a relative position with the train during the push operation and fly alongside it. It receives flight control commands from the cooperative control subsystem and executes flight according to the flight control commands. It collects image data and ranging data of the train and track environment, and collects measurement data of vehicle connection components. It uses a gripper structure to fix or separate itself from the train body. The collaborative control subsystem, connected to the UAV subsystem, train control system, and ground command subsystem, processes the image data, ranging data, and train operation data to generate safety monitoring results. Based on these results, it generates control commands for train operation status and flight control commands, transmitting the control commands to the train control system and the flight control commands to the UAV subsystem. The collaborative control subsystem also processes measurement data of vehicle connection components to generate vehicle connection component location information, sending this information to the operation execution subsystem. Furthermore, the collaborative control subsystem receives path adjustment commands from the ground command subsystem, updates the control commands for the train control system and / or the UAV subsystem based on these commands, and controls the train and / or the UAV based on the updated commands. The ground command subsystem, connected to the collaborative control subsystem, is used to generate a path plan for the push operation, adjust the path plan based on obstacle and path deviation information in the safety monitoring results to generate a path adjustment command, and send the path adjustment command to the collaborative control subsystem. The operation execution subsystem, connected to the collaborative control subsystem, is used to move to the target position based on the position information of the vehicle connection components to perform mechanical connection or disconnection operations between vehicles.

2. The cooperative UAV escort system for rail transit push operations according to claim 1, characterized in that, It also includes a deployment adaptation subsystem, which provides takeoff, landing, storage and charging support for the UAV subsystem, including trackside UAV nests located beside the track and vehicle-mounted UAV nests located at the head of the train.

3. The cooperative UAV escort system for rail transit push operations according to claim 1, characterized in that, The drone subsystem is equipped with a gripping structure below it. The gripping structure includes a retractable robotic arm and an adsorption and locking component at the end of the robotic arm, which is used to temporarily fix the drone subsystem to the train body.

4. The cooperative UAV escort system for rail transit pushing operations according to claim 1, characterized in that, The safety monitoring results include distance information between the rear of the train and the target calculated based on the image data and ranging data. The process by which the collaborative control subsystem processes the image data, ranging data, and train operation data to generate safety monitoring results includes: The image data and the ranging data are timestamped to obtain synchronized image data and ranging data. The synchronized ranging data is filtered, and the validity of the processed point cloud data is determined. The synchronized image data is enhanced and target recognition is performed to obtain an image recognition result containing target pixel coordinates and recognition confidence, and it is determined whether the recognition confidence is lower than a set threshold. If the processed point cloud data is valid and the recognition confidence reaches the set threshold, then the fusion weight of the image data and the ranging data is determined according to the environmental conditions, the target pixel coordinates are converted into three-dimensional spatial coordinates, and the converted three-dimensional spatial coordinates are registered with the processed point cloud data; according to the fusion weight, the registered three-dimensional spatial coordinates and the registered point cloud data are weighted and fused to obtain the first target position data. If the processed point cloud data is invalid or the recognition confidence level does not reach the set threshold, then according to the type of invalid data, historical ranging data or historical image data is used for smooth prediction to obtain the second target location data. Based on the first target location data or the second target location data, as well as the UAV positioning data and train speed information, the relative distance between the train tail end and the target is calculated and motion compensation is performed to generate the distance information in the safety monitoring results.

5. The cooperative UAV escort system for rail transit pushing operations according to claim 4, characterized in that, The collaborative control subsystem is used to compare the distance information in the safety monitoring results with multiple preset graded safety distance thresholds; When the distance information reaches different levels of safety distance thresholds, the collaborative control subsystem generates corresponding level prompt information and sends it to the driver and the ground command subsystem; The collaborative control subsystem is also used to generate a train speed control command corresponding to the preset graded safety distance threshold when the distance information reaches the preset graded safety distance threshold. The speed control command is used to control the train to reduce to a safe speed that matches the current distance.

6. The cooperative UAV escort system for rail transit pushing operations according to claim 2, characterized in that, The ground command subsystem is also used to select the trackside deployment mode or the vehicle-mounted deployment mode of the machine nest according to the push operation route scenario. For fixed-line operations, the trackside deployment mode of the machine nest is preferred, and the nearest trackside machine nest is assigned as the take-off and landing point; for mobile push operations, temporary dedicated line operations, or emergency rescue operations, the vehicle-mounted deployment mode of the machine nest is selected. In mixed-scenario operations involving both fixed and temporary lines, the automatic switching between the trackside deployment mode and the vehicle-mounted deployment mode of the drone nest is controlled: when the train reaches the mode switching node, a mode switching command and target drone nest location information are sent to the UAV subsystem, controlling the UAV subsystem to switch control links, synchronize mission parameters, and fly to the target drone nest to complete landing or anchoring.

7. The cooperative UAV escort system for rail transit pushing operations according to claim 1, characterized in that, The unmanned aerial vehicle (UAV) subsystem is used to scan the route and surrounding area in real time and detect obstacles. The collaborative control subsystem is used to send alarm information containing the location and type of obstacle to the driver and the ground command subsystem when an obstacle is detected. The cooperative control subsystem is also used to generate control commands for the train's operating status based on the threat level of the obstacle, including deceleration commands or emergency stop commands. The collaborative control subsystem is also used to generate obstacle avoidance commands for the UAV based on the location of the obstacle and send them to the UAV subsystem to control the UAV subsystem to avoid the obstacle.

8. The cooperative UAV escort system for rail transit push operations according to claim 1, characterized in that, The vehicle connection components include a vehicle hook and an air duct interface; When the train is pushed to the designated work position, the UAV subsystem is used to locate, identify and obtain the three-dimensional coordinates of the vehicle hook and the air duct interface by fusion positioning of camera and lidar; The collaborative control subsystem is used to receive the three-dimensional coordinates and forward the three-dimensional coordinates to the job execution subsystem; The operation execution subsystem includes a grouped robot and an operation robot dog; the grouped robot is used to move to the hook position according to the three-dimensional coordinates of the vehicle hook and perform hook connection or disconnection operations; the operation robot dog is used to move to the interface position according to the three-dimensional coordinates of the duct interface and simultaneously perform duct docking, disassembly and sealing detection operations. The unmanned aerial vehicle subsystem is also used to monitor the grouped robots and the operational robot dog during the operation, and to feed back the operation footage to the ground command subsystem. The ground command subsystem is used to determine whether there is a deviation in the operation based on the operation screen, and to generate adjustment instructions when a deviation occurs.

9. The cooperative UAV escort system for rail transit pushing operations according to claim 1, characterized in that, If the communication interruption time between the UAV subsystem and the cooperative control subsystem or the ground command subsystem reaches a first threshold, the UAV subsystem is controlled to fly along a preset path and attempt to reconnect; if the communication interruption time exceeds a second threshold, the UAV subsystem is controlled to automatically return to the nearest nest or safe area to land. The unmanned aerial vehicle (UAV) subsystem is used to monitor its remaining battery power in real time and automatically trigger a return-to-home command when the remaining battery power is lower than the return-to-home threshold, controlling itself to return to the nest. If the operation execution subsystem fails, the UAV subsystem is controlled to re-collect and send the coordinate data of the vehicle connection components to guide the operation execution subsystem to retry the operation; if the number of failed retries reaches a set number, an alarm message is sent to the ground command subsystem.

10. A method for drone escort, characterized in that, Using the cooperative unmanned aerial vehicle (UAV) escort system for rail transit push operations as described in any one of claims 1 to 9, the method includes: The unmanned aerial vehicle (UAV) subsystem takes off from its designated nest and arrives near the train based on the push operation task issued by the ground command subsystem; The unmanned aerial vehicle (UAV) subsystem maintains a relative position with the train and flies alongside it, continuously collecting the image data and ranging data during the flight. The collaborative control subsystem processes the image data, ranging data, and train operation data to generate safety monitoring results. Based on the safety monitoring results, it generates control commands for the train operation status and flight control commands. The control commands for the train operation status are transmitted to the train control system, and the flight control commands are transmitted to the UAV subsystem. The ground command subsystem adjusts the path planning based on the obstacle and path deviation information in the safety monitoring results, generates a path adjustment command, and sends it to the collaborative control subsystem; the collaborative control subsystem updates the control commands to the train control system and / or the UAV subsystem based on the path adjustment command, and controls the train and / or UAV based on the updated control commands; When the train is pushed to the designated position, the UAV subsystem collects measurement data of the vehicle connection components; the collaborative control subsystem processes the measurement data of the vehicle connection components to generate vehicle connection component position information and sends it to the operation execution subsystem; the operation execution subsystem moves to the operation position according to the vehicle connection component position information to complete the connection or disconnection operation.

Images