Cooperative distribution method for unmanned aerial vehicle and unmanned vehicle

Through the coordinated work of drones and unmanned vehicles, high-altitude exploration and deep learning models are used to monitor and adjust the distribution routes in real time, solving the problem that unmanned vehicles cannot plan and avoid obstacles in advance, and improving the efficiency and stability of logistics and distribution.

CN119027010AInactive Publication Date: 2024-11-26ZHEJIANG XIANGLONG AVIATION TECH CO LTD

Patent Information

Application Number
CN202411537263.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-31
Publication Date
2024-11-26
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing unmanned vehicles cannot plan their routes in advance during logistics and distribution, and cannot predict and avoid obstacles, resulting in low distribution efficiency and poor stability.

Method used

Using drones and unmanned vehicles to work in concert, use drones to detect routes at high altitudes, use image sensors and deep learning models to monitor road conditions in real time, plan and adjust the distribution routes of unmanned vehicles, and realize obstacle avoidance operations for static and dynamic obstacles. .

Benefits of technology

It improves distribution efficiency and stability, reduces manpower demand, and through multiple iterations to optimize route calculations, reduces the amount of calculations during the road search process, ensures that unmanned vehicles can accurately avoid dynamic obstacles, and improves the automation level of the distribution process.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119027010A_ABST
    Figure CN119027010A_ABST
Patent Text Reader

Abstract

The invention belongs to the technical field of logistics transportation, and relates to an unmanned aerial vehicle and unmanned vehicle collaborative distribution method. In the logistics distribution process, a distribution route planned by the unmanned vehicle in advance cannot pass through due to obstacles on the road, and the unmanned vehicle can complete obstacle avoidance operation only after approaching. According to the invention, through cooperative work of the unmanned aerial vehicle and the unmanned vehicle, the unmanned aerial vehicle carries out multiple iterations through routes among a plurality of operators, searches for a distribution route for the unmanned vehicle, and carries out iteration on the position between the unmanned vehicle and the dynamic obstacle, so that the unmanned vehicle and the dynamic obstacle in the space are recombined, and the unmanned aerial vehicle distribution is realized. And the time difference between the two arrives at the weight and the place is fed back, so that the distribution process is automatic, and the distribution efficiency and the stability in the distribution process are remarkably improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of logistics and transportation, and relates to a method for collaborative delivery between a drone and an unmanned vehicle. Background Art

[0002] The work that can be done in two dimensions, that is, on the ground, is limited. So using drones to complete some work in three dimensions can not only relieve congestion in two dimensions, but also make full use of space resources. In daily work, it breaks through the limitations of vision and greatly facilitates people's work and life.

[0003] In the existing technology, unmanned vehicles have been widely used in the field of logistics and distribution. However, in the process of logistics and distribution, the routes of unmanned vehicles cannot be planned in advance, or the delivery routes planned in advance by unmanned vehicles cannot pass due to obstacles on the road. At the same time, unmanned vehicles cannot predict the obstacles they will encounter and cannot complete the obstacle avoidance operation in advance. They can only complete the obstacle avoidance operation after approaching. Summary of the invention

[0004] In order to solve the problems existing in the background technology, the present invention proposes a collaborative delivery method of a drone and an unmanned vehicle.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is as follows: A method for collaborative delivery between a drone and an unmanned vehicle, comprising: The user inputs the delivery location and assigns drones and unmanned vehicle teams to carry out the delivery task; Before drones and unmanned vehicle teams perform delivery tasks, multiple delivery locations are marked and mapped on a two-dimensional map; Before the unmanned vehicle performs a delivery mission, the drone will explore all available routes at high altitude and plan the delivery route for the unmanned vehicle; The unmanned vehicle performs the delivery task according to the route planned by the drone. During the driving process of the unmanned vehicle, the unmanned vehicle autonomously completes the static obstacle avoidance operation; The drone uses image sensors to collect data during the delivery mission, and the data collected by the drone is submitted to the deep learning model as parameter input; When the unmanned vehicle is performing a delivery task, it uses the instructions sent by the drone to complete dynamic obstacle avoidance operations; When unmanned vehicles and drones coordinate to perform delivery tasks, the deep learning model monitors the road conditions of the current delivery route in real time based on the data collected by the drones, and adjusts the unmanned vehicle delivery route in real time based on the road conditions.

[0006] Furthermore, the specific method for the drone to plan the delivery route for the unmanned vehicle is: The drone extracts the distance between every two delivery locations and generates a distance matrix based on the distance between every two delivery locations; The drone creates several operators, each of which generates any combination of delivery routes, and each operator retains the distance between every two delivery locations, that is, each operator retains the total distance required for the corresponding delivery route; The drone controls any two operators to exchange the connection relationship between any delivery locations. Each operator compares the distance after the exchange with the distance before the exchange. If the distance is shorter, the exchange is retained; if the distance is longer, the exchange is cancelled. Iterate between operators until the maximum number of iterations is reached or the distance does not change after the preset number of iterations. The delivery route corresponding to the operator with the shortest total distance is retained, and the delivery route with the shortest total distance is sent to the unmanned vehicle.

[0007] Furthermore, during the driving process of the unmanned vehicle, the specific method for the unmanned vehicle to complete the static obstacle avoidance operation is: The unmanned vehicle uses an image sensor to obtain images of the road ahead and analyzes static obstacles in the direction of the unmanned vehicle based on the road images; Setting safe distance thresholds ,If the distance of the detected static obstacle is less than the threshold, it is considered that there is a potential collision risk; Obtain the distance between the static obstacle and the unmanned vehicle based on the size of the static obstacle in the road image and the focal length of the image sensor , the specific formula is:

[0008] in is the actual height of the static obstacle, is the focal length of the image sensor, is the pixel height of the static obstacle in the road image, and the distance between the static obstacle and the unmanned vehicle is calculated. If the distance between the static obstacle and the unmanned vehicle Less than the safety distance threshold , then the unmanned vehicle is controlled to turn; The unmanned vehicle uses its own forward direction as the dividing line to segment the static obstacles, and determines the size of the static obstacles on both sides of the dividing line in the road image. The unmanned vehicle deflects to the side with a smaller static obstacle area. After the unmanned vehicle passes the static obstacle, it deflects in the opposite direction according to the deflection angle and returns to the normal driving route.

[0009] Furthermore, during the unmanned vehicle's delivery mission, the specific method of using the instructions sent by the drone to complete the dynamic obstacle avoidance operation is: The drone records the real-time changes in the position of dynamic obstacles, determines the moving direction and speed of dynamic obstacles, and predicts the moving position of dynamic obstacles; The drone obtains the current location of the unmanned vehicle and determines the location of the unmanned vehicle based on its current location and speed. Position after time; The UAV merges the moving path of the dynamic obstacle with the moving path of the unmanned vehicle, and determines whether the moving path of the dynamic obstacle intersects with the moving path of the unmanned vehicle; If no intersection occurs, no intervention is made. If an intersection occurs, a Petri model is established based on the driving path of the unmanned vehicle and the moving path of the dynamic obstacle, and the unmanned vehicle avoids the dynamic obstacle based on the Petri model.

[0010] Furthermore, the specific method of establishing the Petri model is: Will Set to the starting position of the unmanned vehicle and the initial speed of the unmanned vehicle ,Will Set as the starting position of the dynamic obstacle and the initial speed of the dynamic obstacle ; Mobile for driverless cars The location after time, Move for dynamic obstacles Location after time; According to the moving path of the dynamic obstacle and the moving path of the unmanned vehicle, the starting position and initial speed of the unmanned vehicle, the starting position and initial speed of the dynamic obstacle, the dynamic obstacle is iterated. and position until and Overlap, set the time required for the unmanned vehicle to reach the intersection as , the time required for the dynamic obstacle to reach the intersection is ,judge and The time difference between the two, if the time difference is less than the minimum response time, the unmanned vehicle is controlled to slow down; if the time difference is greater than or equal to the minimum response time, the unmanned vehicle drives normally.

[0011] Furthermore, the prediction of the position of the unmanned vehicle by the drone includes the path deviation generated when the unmanned vehicle avoids static obstacles.

[0012] Furthermore, the specific method for adjusting the unmanned vehicle delivery route in real time based on road conditions is: The drone deletes the delivery points that the unmanned vehicle has delivered and performs the route planning operation again; Delete the current problematic road from the two-dimensional map, modify the route distance from the current position of the unmanned vehicle to the next delivery point, and replan the route. Compared with the prior art, the present invention has the following beneficial effects: The present invention realizes automation of the delivery process through the collaborative work of drones and unmanned vehicles, reduces the need for manpower, and significantly improves the delivery efficiency. The present invention performs multiple iterations of routes between multiple operators, greatly improving the delivery efficiency of drones and unmanned vehicles, and greatly reducing the amount of calculation in the path-finding process.

[0013] The present invention iterates the position between the unmanned vehicle and the dynamic obstacle, so that the unmanned vehicle and the dynamic obstacle are reconciled in space, and the time difference between the two arriving at the reconciled location is fed back, thereby accurately achieving the unmanned vehicle's avoidance efficiency in the face of dynamic obstacles during the delivery process and improving the stability of the delivery process. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0015] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments 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 creative work are within the scope of protection of the present invention.

[0016] like Figure 1 As shown, the technical solution adopted by the present invention is as follows: a method for collaborative delivery between a drone and an unmanned vehicle, comprising: The user inputs the delivery location and assigns drones and unmanned vehicle teams to carry out delivery tasks.

[0017] Before drones and unmanned vehicles perform delivery tasks, multiple delivery locations are marked and mapped on a two-dimensional map.

[0018] Before the unmanned vehicle performs a delivery mission, the drone explores all available routes at high altitudes and plans the delivery route for the unmanned vehicle.

[0019] The unmanned vehicle performs delivery tasks according to the route planned by the drone. During the driving process of the unmanned vehicle, the unmanned vehicle independently completes static obstacle avoidance operations.

[0020] When the drone performs a delivery mission, it uses image sensors to collect data, and the data collected by the drone is submitted to the deep learning model as parameter input.

[0021] When the unmanned vehicle is performing a delivery mission, it uses the instructions sent by the drone to complete dynamic obstacle avoidance operations.

[0022] When unmanned vehicles and drones coordinate to perform delivery tasks, the deep learning model monitors the road conditions of the current delivery route in real time based on the data collected by the drones, and adjusts the unmanned vehicle delivery route in real time based on the road conditions.

[0023] When a user has a delivery demand, the user enters multiple delivery locations in the system and assigns drones and unmanned vehicle teams to perform delivery tasks.

[0024] Before drones and unmanned vehicles perform delivery tasks, multiple delivery locations are marked and mapped on a two-dimensional map.

[0025] The drone extracts the distance between each two delivery locations and generates a distance matrix. Generating distances between each two delivery locations ensures that the drone takes all possible routes into account when planning routes for the unmanned vehicle, avoiding the loss of the shortest possible route due to the lack of marking of a route.

[0026] The drone creates several operators, each of which generates any combination of delivery routes, and each operator retains the distance between every two delivery locations, that is, each operator retains the total distance required for the corresponding delivery route. Multiple operators are used to randomly plan routes, and the delivery route represented by each operator includes all delivery locations, and the starting point and end point are consistent, and there are no repeated delivery points in the delivery route.

[0027] The drone controls any two operators to exchange the connection relationship between any delivery locations. Each operator compares the distance after the exchange with the distance before the exchange. If the distance is shorter, the exchange is retained; if the distance is longer, the exchange is cancelled. It can ensure that each operator can be replaced with a shorter route relative to the current route, so that the delivery route represented by each operator can be continuously shortened in multiple iterations.

[0028] Operators are iterated until the maximum number of iterations is reached or the distance does not change after the preset number of iterations. The delivery routes corresponding to the preset number of operators with the shortest total distance are retained, and the delivery route with the shortest total distance is sent to the unmanned vehicle. Retaining the delivery routes corresponding to multiple operators can be replaced when the shortest delivery route is not passable without recalculation.

[0029] After the delivery route calculation is completed, the unmanned vehicle performs the delivery task according to the route planned by the drone. During the driving process of the unmanned vehicle, the unmanned vehicle autonomously completes the static obstacle avoidance operation.

[0030] The unmanned vehicle uses an image sensor to obtain images of the road ahead and analyzes static obstacles in the direction of the unmanned vehicle based on the road images. The unmanned vehicle can extract obstacles using gradient values ​​or edge extraction operations.

[0031] Setting safe distance thresholds If the distance of the detected static obstacle is less than the threshold, it is considered that there is a potential collision risk. In order to prevent the unmanned vehicle from deviating from the route or increasing energy consumption during driving, the unmanned vehicle only avoids static obstacles within a safe distance.

[0032] Obtain the distance between the static obstacle and the unmanned vehicle based on the size of the static obstacle in the road image and the focal length of the image sensor , the specific formula is:

[0033] in is the actual height of the static obstacle, is the focal length of the image sensor, is the pixel height of the static obstacle in the road image, and the distance between the static obstacle and the unmanned vehicle is calculated. If the distance between the static obstacle and the unmanned vehicle Less than the safety distance threshold , then the unmanned vehicle is controlled to turn.

[0034] The unmanned vehicle uses its own forward direction as the dividing line to segment the static obstacles, and determines the size of the static obstacles on both sides of the dividing line in the road image. The unmanned vehicle deflects to the side with a smaller static obstacle area. After the unmanned vehicle passes the static obstacle, it deflects in the opposite direction according to the deflection angle and returns to the normal driving route. The unmanned vehicle turns to avoid the side with a smaller static obstacle area, which can reduce the consumption of the unmanned vehicle during driving. At the same time, after passing the obstacle, the unmanned vehicle deflects in the opposite direction according to the deflection angle, which does not interfere with the normal driving of the unmanned vehicle, ensuring that the unmanned vehicle can still drive normally after avoiding it.

[0035] When the unmanned vehicle performs the delivery task, it is inevitable to encounter dynamic objects such as pedestrians or cars. The unmanned vehicle cannot complete the avoidance autonomously, and needs the cooperation of the drone to complete the avoidance of dynamic obstacles. The specific method of using the instructions sent by the drone to complete the dynamic obstacle avoidance operation is: The drone records the real-time changes in the position of dynamic obstacles, determines the moving direction and speed of dynamic obstacles, and predicts the moving position of dynamic obstacles. The drone predicts the movement of dynamic obstacles based on the current speed and direction of dynamic obstacles.

[0036] The drone obtains the current location of the unmanned vehicle and determines the location of the unmanned vehicle based on its current location and speed. Position after time. The size can be set freely by the user. The larger the value, the more stable the unmanned vehicle will be in avoiding dynamic obstacles during driving, and the more stable the unmanned vehicle delivery process will be.

[0037] The prediction of the position of the unmanned vehicle by the drone includes the path deviation caused by the unmanned vehicle avoiding static obstacles, so as to avoid subsequent calculation errors caused by the unmanned vehicle avoiding static obstacles.

[0038] The drone merges the moving path of the dynamic obstacle with the moving path of the unmanned vehicle, and determines whether the moving path of the dynamic obstacle intersects with the moving path of the unmanned vehicle. The merging operation of the moving path of the dynamic obstacle and the moving path of the unmanned vehicle can be completed through two-dimensional spatial topology.

[0039] If no intersection occurs, no intervention is made. If an intersection occurs, a Petri model is established based on the driving path of the unmanned vehicle and the moving path of the dynamic obstacle, and the unmanned vehicle avoids the dynamic obstacle based on the Petri model. If no intersection occurs, there is no related expansion between the unmanned vehicle and the dynamic obstacle. If an intersection occurs, there is a probability that the unmanned vehicle and the dynamic obstacle will collide, and the Petri model needs to be used for calculation.

[0040] The Petri model includes: Will Set to the starting position of the unmanned vehicle and the initial speed of the unmanned vehicle ,Will Set as the starting position of the dynamic obstacle and the initial speed of the dynamic obstacle .

[0041] Mobile for driverless cars The location after time, Move for dynamic obstacles The location after time.

[0042] According to the moving path of the dynamic obstacle and the moving path of the unmanned vehicle, the starting position and initial speed of the unmanned vehicle, and the starting position and initial speed of the dynamic obstacle. and position until and Overlap. and The time dimension is different. Assume that in this embodiment, the distances to the intersection point are the same and the speeds are different. and If they overlap, the time required is different, so and Because they are in different time dimensions, there will be a time difference between the two.

[0043] Set the time required for the unmanned vehicle to reach the intersection as , the time required for the dynamic obstacle to reach the intersection is ,judge and If the time difference between the two is less than the minimum response time, the unmanned vehicle is controlled to slow down. If the time difference is greater than or equal to the minimum response time, the unmanned vehicle drives normally. This ensures that when the unmanned vehicle faces a dynamic obstacle with a collision risk, it can predict in advance whether it will collide with the dynamic obstacle and respond to the dynamic obstacle in advance.

[0044] If an extreme situation occurs and the unmanned vehicle cannot pass a certain section of the road, the unmanned vehicle delivery route will be adjusted in real time based on the road conditions. The specific method is as follows: The drone deletes the delivery points that have been delivered by the unmanned vehicle and performs the route planning operation again.

[0045] Delete the current problematic road from the two-dimensional map, modify the route distance from the current location of the unmanned vehicle to the next delivery point, and replan the route. In this way, the unmanned vehicle can still complete the delivery task in extreme cases.

[0046] Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments, or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for collaborative delivery between drones and unmanned vehicles. It is characterized in that Included are: The user inputs the delivery location and assigns drones and unmanned vehicle teams to carry out the delivery task; Before drones and unmanned vehicle teams perform delivery tasks, multiple delivery locations are marked and mapped on a two-dimensional map; Before the unmanned vehicle performs a delivery mission, the drone will explore all available routes at high altitude and plan the delivery route for the unmanned vehicle; The unmanned vehicle performs the delivery task according to the route planned by the drone. During the driving process of the unmanned vehicle, the unmanned vehicle autonomously completes the static obstacle avoidance operation; The drone uses image sensors to collect data during the delivery mission, and the data collected by the drone is submitted to the deep learning model as parameter input; When the unmanned vehicle is performing a delivery task, it uses the instructions sent by the drone to complete dynamic obstacle avoidance operations; When unmanned vehicles and drones coordinate to perform delivery tasks, the deep learning model monitors the road conditions of the current delivery route in real time based on the data collected by the drones, and adjusts the unmanned vehicle delivery route in real time based on the road conditions.

2. A method for coordinated delivery of unmanned aerial vehicles and unmanned vehicles according to claim 1, It is characterized in that The specific method for the drone to plan the delivery route for the unmanned vehicle is: The drone extracts the distance between every two delivery locations and generates a distance matrix based on the distance between every two delivery locations; The drone creates several operators, each of which generates any combination of delivery routes, and each operator retains the distance between every two delivery locations, that is, each operator retains the total distance required for the corresponding delivery route; The drone controls any two operators to exchange the connection relationship between any delivery locations. Each operator compares the distance after the exchange with the distance before the exchange. If the distance is shorter, the exchange is retained; if the distance is longer, the exchange is cancelled. Iterate between operators until the maximum number of iterations is reached or the distance does not change after the preset number of iterations. The delivery route corresponding to the operator with the shortest total distance is retained, and the delivery route with the shortest total distance is sent to the unmanned vehicle.

3. A method for coordinated delivery of unmanned aerial vehicles and unmanned vehicles according to claim 2, It is characterized in that The specific method for the unmanned vehicle to complete the static obstacle avoidance operation during the driving process of the unmanned vehicle is: The unmanned vehicle uses an image sensor to obtain images of the road ahead and analyzes static obstacles in the direction of the unmanned vehicle based on the road images; Setting safe distance thresholds ,If the distance of the detected static obstacle is less than the threshold, it is considered that there is a potential collision risk; Obtain the distance between the static obstacle and the unmanned vehicle based on the size of the static obstacle in the road image and the focal length of the image sensor , the specific formula is: in is the actual height of the static obstacle, is the focal length of the image sensor, is the pixel height of the static obstacle in the road image, and the distance between the static obstacle and the unmanned vehicle is calculated. If the distance between the static obstacle and the unmanned vehicle Less than the safety distance threshold , then the unmanned vehicle is controlled to turn; The unmanned vehicle uses its own forward direction as the dividing line to segment the static obstacles, and determines the size of the static obstacles on both sides of the dividing line in the road image. The unmanned vehicle deflects to the side with a smaller static obstacle area. After the unmanned vehicle passes the static obstacle, it deflects in the opposite direction according to the deflection angle and returns to the normal driving route.

4. A drone and unmanned vehicle collaborative delivery method according to claim 3, It is characterized in that When the unmanned vehicle performs a delivery task, the specific method of using the command sent by the drone to complete the dynamic obstacle avoidance operation is as follows: The drone records the real-time changes in the position of dynamic obstacles, determines the moving direction and speed of dynamic obstacles, and predicts the moving position of dynamic obstacles; The drone obtains the current location of the unmanned vehicle and determines the location of the unmanned vehicle based on its current location and speed. Position after time; The UAV merges the moving path of the dynamic obstacle with the moving path of the unmanned vehicle, and determines whether the moving path of the dynamic obstacle intersects with the moving path of the unmanned vehicle; If no intersection occurs, no intervention is made. If an intersection occurs, a Petri model is established based on the driving path of the unmanned vehicle and the moving path of the dynamic obstacle, and the unmanned vehicle avoids the dynamic obstacle based on the Petri model.

5. A method for coordinated delivery of drones and unmanned vehicles according to claim 4, It is characterized in that The specific method of establishing the Petri model is: Will Set to the starting position of the unmanned vehicle and the initial speed of the unmanned vehicle ,Will Set as the starting position of the dynamic obstacle and the initial speed of the dynamic obstacle ; Mobile for driverless cars The location after time, Move for dynamic obstacles Location after time; According to the moving path of the dynamic obstacle and the moving path of the unmanned vehicle, the starting position and initial speed of the unmanned vehicle, the starting position and initial speed of the dynamic obstacle, the dynamic obstacle is iterated. and position until and Overlap, set the time required for the unmanned vehicle to reach the intersection as , the time required for the dynamic obstacle to reach the intersection is ,judge and The time difference between the two, if the time difference is less than the minimum response time, the unmanned vehicle is controlled to slow down; if the time difference is greater than or equal to the minimum response time, the unmanned vehicle drives normally.

6. A method for coordinated delivery of unmanned aerial vehicles and unmanned vehicles according to claim 4, It is characterized in that The prediction of the position of the unmanned vehicle by the drone includes the path deviation generated when the unmanned vehicle avoids static obstacles.

7. A method for coordinated delivery of unmanned aerial vehicles and unmanned vehicles according to claim 4, It is characterized in that The specific method for adjusting the unmanned vehicle delivery route in real time based on road conditions is: The drone deletes the delivery points that the unmanned vehicle has delivered and performs the route planning operation again; Delete the current problematic road from the two-dimensional map, modify the route distance from the current location of the unmanned vehicle to the next delivery point, and replan the route.

Citation Information

Patent Citations

  • Air-land unmanned smart decision making method

    CN108229685A

  • Vehicle path optimization method based on Petri network and integer linear programming

    CN111325389A

  • System and method for providing local path planning for ground unmanned vehicle by unmanned aerial vehicle

    CN116540706A

  • Unmanned vehicle tracking control method based on fuzzy Petri and DDPG network

    CN117452936A

  • Air-ground cooperative path planning method based on unmanned aerial vehicle and unmanned vehicle

    CN118443015A

Cited By

  • Air-ground cooperative distribution route optimization method for unmanned aerial vehicle

    CN120447592A