An interaction method and system for an autonomous driving dedicated service vehicle

Abstract

The application relates to the field of automatic driving, in particular to an interaction method and system of an automatic driving special work vehicle. The method comprises the following steps: an automatic driving controller receives a vehicle registration signal and map information to plan a path, controls the vehicle to run to a specified position, sends the path planning to a cloud control platform, and judges in real time whether an obstacle is encountered, if an obstacle is encountered, the vehicle driving mode is switched to a remote takeover mode, if no obstacle is encountered, the vehicle runs to the specified position in an automatic driving mode; when the vehicle reaches the specified position, the cloud control platform sends a work instruction to a work device controller; the work device controller executes the work instruction and identifies work completion, and sends the work completion to the cloud control platform. The application has the beneficial technical effects that the driving stability and work efficiency of the automatic driving special work vehicle are improved through path planning and switching of the driving mode when an obstacle is encountered.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving, and in particular to an interaction method and system for a dedicated autonomous driving operation vehicle. Background Art

[0002] While the autonomous driving capabilities of a single vehicle can handle complex and ever-changing traffic scenarios, the increasing number of intelligent vehicles necessitates the management and coordinated dispatch of multiple vehicles. This is especially true for autonomous dedicated work vehicles, such as unmanned express delivery vehicles, unmanned retail vehicles, and unmanned street sweepers. This requires unified management and dispatch of all autonomous dedicated work vehicles through a cloud-based control platform. The interaction between vehicles and cloud-based control platforms is the core technology that determines the large-scale commercialization of autonomous dedicated work vehicles. However, the current interaction technology between autonomous dedicated work vehicles and cloud-based control platforms suffers from poor generalizability, resulting in technical issues such as low operational efficiency and insufficient driving stability for autonomous dedicated work vehicles. Therefore, it is necessary to develop an interaction method for autonomous dedicated work vehicles that can enhance the generalizability of interaction technology, improve operational efficiency, and enhance driving stability.

[0003] For example, Chinese patent CN114169741A, published on March 11, 2022, is an unmanned express delivery vehicle system. The invention includes a cloud system, an on-board system, an identification system, and a controller. The identification system includes a camera. The cloud system is equipped with an express entry unit and an express distribution unit. The express entry unit obtains basic express information through the camera, generates a receipt summary table and generates route information through the distribution strategy, and the express distribution unit sends the route information to the controller. The identification system also includes a size unit, which is used to obtain the size of the express and generate the loading sequence. The on-board system includes a delivery unit, a feedback unit, and a reset planning unit. The on-board system delivers the express, sends prompt information, and plans the reset route for delivery to the collection point. It has the effect of efficiently delivering express and high delivery accuracy. However, this technical solution does not solve the technical problem of poor driving stability of the dedicated autonomous driving operation vehicle in the prior art. Summary of the Invention

[0004] The technical problem to be solved by the present invention is: the current interaction technology of autonomous driving special-purpose work vehicles has the technical problems of poor universality, low operating efficiency of autonomous driving special-purpose work vehicles, and poor driving stability. An interaction method and system for autonomous driving special-purpose work vehicles are proposed, which improve the driving stability and operating efficiency of autonomous driving special-purpose work vehicles through path planning and switching driving modes when encountering obstacles.

[0005] To solve the above technical problems, the present invention adopts the following technical solution: an interaction method for an autonomous driving dedicated work vehicle, which is used to realize the interaction between an autonomous driving controller, a cloud control platform, and a work device controller, comprising the following steps:

[0006] The autonomous driving controller receives the vehicle registration signal and map information, performs path planning based on the vehicle registration signal and map information, and sends the path planning to the cloud control platform;

[0007] The automatic driving controller controls the vehicle to run to the designated location according to the path planning, and determines in real time whether it encounters an obstacle. If an obstacle is encountered, the vehicle driving mode is switched to the remote takeover mode. If no obstacle is encountered, the vehicle is controlled to run to the designated location in the automatic driving mode.

[0008] When the vehicle arrives at the designated location, the cloud control platform sends an operation instruction to the operation device controller;

[0009] The operation device controller executes the operation instruction and identifies the completion status of the operation, and sends the completion status of the operation to the cloud control platform.

[0010] Preferably, the vehicle registration signal includes a vehicle ID and vehicle location information, the map information includes a map of the area where the vehicle is currently located and designated location information, and the path planning method includes:

[0011] Obtain all candidate navigation paths from the vehicle position to the designated location based on the vehicle position information, a map of the area the vehicle is currently located in, and the designated location information;

[0012] Extracting static data of the road sections of the candidate navigation path;

[0013] Reading the historical road segment dynamic data of the candidate navigation path;

[0014] The cost values ​​of all candidate navigation paths are calculated based on the static data of the road segment and the dynamic data of the historical road segment, and the candidate navigation path with the smallest cost value is used as the vehicle navigation path for this operation.

[0015] Preferably, the method for extracting static data of the road section of the candidate navigation path includes:

[0016] Loading a map of the area where the autonomous driving dedicated operation vehicle is currently located;

[0017] All candidate navigation paths are segmented and lane information for each segment is extracted. The lane information includes segment length l, segment width W, segment flatness P, number of intersections I, and road curvature C. The length L and number of segments N of all candidate navigation paths are obtained.

[0018] The lane information, the length of the candidate navigation path, and the number of road sections constitute road section static data.

[0019] Preferably, the historical road section dynamic data includes the obstacle flow F, congestion probability G and congestion degree E of the road section, and the method for calculating the cost value of the candidate navigation path includes:

[0020] The length cost of the path: L cost =L×S L ,

[0021] Intersection quantity cost: I cost =I×S I ,

[0022] The width cost of the path: W cost =S W ÷(W1+W2+…+W N ),

[0023] The flatness cost of the path: P cost =S P ÷(P1+P2+…+P N ),

[0024] The path curvature cost value: C cost =(C1+C2+…+S N )×S C ,

[0025] Obstacle flow cost of the path: F cost =(F1+F2+…+F N )×D F ,

[0026] Congestion probability and degree cost of the path: J cost =(G1×E1+G2×E2+…+G N ×E N )×D J ,

[0027] The total cost of the candidate navigation path is:

[0028] T cost =L cost +I cost +W cost +P cost +C cost +F cost +J cost

[0029] Among them, S L is the length cost coefficient of the preset path, S Iis the preset intersection number cost coefficient, S W is the width cost coefficient of the preset path, S P is the flatness cost coefficient of the preset path, S C is the road curvature cost coefficient of the preset path, D F is the obstacle flow cost coefficient of the preset path, D J It is the congestion probability and degree cost coefficient of the preset path.

[0030] Preferably, the obstacle flow F includes motor vehicle flow V, non-motor vehicle flow B, pedestrian flow X, and unknown obstacle flow U. The method for calculating the obstacle flow cost value of the candidate navigation path further includes:

[0031] The cost of vehicle traffic flow on the path: V cost =(V1+V2+…+V N )×D V ,

[0032] Non-motorized vehicle flow cost value of the path: B cost =(B1+B2+…+B N )×D B ,

[0033] Pedestrian flow cost value of the path: X cost =(X1+X2+…+X N )×S X ,

[0034] The unknown obstacle flow cost of the path: U cost =(U1+U2+…+U N )×D U ,

[0035] The obstacle flow cost of the candidate navigation path is:

[0036] F cost =V cost +B cost +X cost +U cost

[0037] Among them, D V is the motor vehicle flow cost coefficient of the preset path, D B is the non-motor vehicle flow cost coefficient of the preset path, D X is the pedestrian flow cost coefficient of the preset path, D U is the flow cost coefficient of the unknown obstacle in the preset path.

[0038] Preferably, the method for determining whether an obstacle is encountered includes:

[0039] The autonomous driving controller updates the vehicle’s positioning information in real time during operation;

[0040] If the positioning information does not change after exceeding a preset time threshold, it is determined that the vehicle has encountered an obstacle. If the positioning information changes in real time, it is determined that the vehicle has not encountered an obstacle.

[0041] Preferably, the method for switching the vehicle driving mode to the remote takeover mode after determining that the vehicle encounters an obstacle includes:

[0042] When the autonomous driving controller determines that the vehicle encounters an obstacle and the autonomous driving mode cannot bypass it, it reports to the cloud control platform and switches the vehicle driving mode to the remote takeover mode;

[0043] The automatic driving controller receives and executes the vehicle movement remote control instruction sent by the remote duty personnel to the vehicle until the vehicle bypasses the obstacle.

[0044] Preferably, the cloud control platform records the detour starting point and detour path in the remote takeover mode;

[0045] Segmenting the multiple detour paths recorded at the same detour starting point, making each segment a straight line segment or a circular arc segment, recording the direction and length of the straight line segment, and recording the radius and radian of the circular arc segment;

[0046] Setting a similarity threshold, treating two straight line segments whose similarity exceeds the similarity threshold as similar segments, and treating two arc segments whose similarity exceeds the similarity threshold as similar segments;

[0047] Counting the number of similar segments of each straight line segment and arc segment, and recording the straight line segment and arc segment whose number of similar segments exceeds a preset threshold as the preferred detour operation;

[0048] When the subsequent vehicle runs to the detour starting point again, the automatic driving controller determines whether the obstacle still exists. If the obstacle still exists, the automatic driving controller attempts to detour according to the preferred detour operation. If the obstacle disappears, the vehicle runs to the designated position in the automatic driving mode.

[0049] Preferably, the method of attempting to bypass includes:

[0050] Execute any of the preferred detour operations until there are no obstacles when re-planning the path or the preferred detour operation is executed up to a preset upper limit. If the preferred detour operation is executed up to a preset upper limit, report to the cloud control platform and switch the vehicle driving mode to the remote takeover mode.

[0051] An interactive system for an autonomous driving dedicated work vehicle, comprising:

[0052] an autonomous driving controller, configured to receive a vehicle registration signal and map information, perform path planning based on the vehicle registration signal and map information, control the vehicle to a designated location based on the path planning, and determine in real time whether an obstacle is encountered. If an obstacle is encountered, the vehicle driving mode is switched to a remote takeover mode; if no obstacle is encountered, the vehicle is controlled to operate in the autonomous driving mode to the designated location;

[0053] The cloud control platform is used to send operation instructions to the operation device controller when the vehicle reaches the designated location;

[0054] An operating device controller, configured to execute the operating instructions and identify the completion status of the operating;

[0055] The vehicle-mounted T-BOX is used to achieve communication and transmission between the autonomous driving controller, the operating device controller and the cloud control platform.

[0056] The beneficial technical effects of the present invention include: employing an interaction method and system for an autonomous driving dedicated work vehicle, which improves the generalization of the interaction technology between the autonomous driving dedicated work vehicle and the cloud control platform by clearly describing the interaction process and interaction data content between the autonomous driving dedicated work vehicle and the cloud control platform; improving the driving stability and operating efficiency of the autonomous driving dedicated work vehicle by calculating the cost value of a candidate navigation path for path planning and switching driving modes when encountering obstacles, thereby better achieving functions such as complete autonomous driving operation tasks, vehicle control and scheduling, and remote takeover mode operation when encountering obstacles; utilizing the cloud control platform to perform segmented analysis and similarity calculation on the recorded detour starting points and detour paths in the remote takeover mode, thereby planning more efficient and reasonable detour routes for the autonomous driving dedicated work vehicle when encountering obstacles; and by presetting an upper limit on the number of times the preferred detour operation can be executed and reporting to the cloud control platform to switch the vehicle's driving mode when the preset number of times is exceeded, timely adjustment of the obstacle detour route can be achieved, thereby successfully completing the operation mission of the autonomous driving dedicated work vehicle.

[0057] Other features and advantages of the present invention will be disclosed in detail in the following specific embodiments and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] The present invention will be further described below with reference to the accompanying drawings:

[0059] Figure 1 This is a flow chart of the interaction method for an autonomous driving dedicated work vehicle according to an embodiment of the present invention.

[0060] Figure 2 This is a flow chart of a path planning method according to an embodiment of the present invention.

[0061] Figure 3 This is a schematic diagram of a map of the delivery area of ​​an unmanned express delivery vehicle according to an embodiment of the present invention.

[0062] Figure 4 This is a schematic diagram of the path planning of an unmanned delivery vehicle according to an embodiment of the present invention.

[0063] Figure 5 This is a schematic diagram of an unmanned delivery vehicle automatically driving to the next location according to an embodiment of the present invention.

[0064] Figure 6 This is a schematic diagram of an unmanned delivery vehicle automatically returning to its initial delivery point according to an embodiment of the present invention.

[0065] Figure 7 Schematic diagram of the interactive system structure of the autonomous driving dedicated work vehicle according to an embodiment of the present invention.

[0066] Among them: 1. Automatic driving controller, 2. Cloud control platform, 3. Operating device controller, 4. Vehicle-mounted T-BOX. DETAILED DESCRIPTION

[0067] The following is an explanation and description of the technical solutions of the embodiments of the present invention in conjunction with the drawings of the embodiments of the present invention. However, the following embodiments are only preferred embodiments of the present invention and are not exhaustive. Based on the embodiments in the implementation manner, other embodiments obtained by those skilled in the art without creative work are all within the scope of protection of the present invention.

[0068] In the following description, terms such as "inside", "outside", "up", "down", "left", "right", etc. that indicate directions or positional relationships are only used to facilitate the description of the embodiments and simplify the description, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operate in a specific direction. Therefore, they should not be understood as limiting the present invention.

[0069] Before describing the technical solution of this embodiment in detail, the background of the application of this embodiment is first introduced.

[0070] With the rapid development of autonomous driving technology, the autonomous driving capabilities of intelligent vehicles are becoming increasingly mature. Autonomous driving technology primarily utilizes positioning systems, internet technologies, intelligent technologies, and image processing technologies to achieve unmanned vehicle control by collecting, organizing, and analyzing road conditions and driving information. Furthermore, autonomous driving technology utilizes radar and monitoring systems to transmit vehicle-related information to the vehicle control system, which then effectively completes relevant instructions through scientific and professional calculations. Because infrastructure development lags behind the development of intelligent vehicles, the widespread adoption of autonomous driving technology remains somewhat challenging. However, in specific applications, such as container transportation and express parcel delivery, autonomous driving technology has already been applied, transforming traditional manual operations and improving both efficiency and quality.

[0071] While the autonomous driving capabilities of a single vehicle can handle complex and ever-changing traffic scenarios, there's still significant room for improvement as the number of intelligent vehicles increases. Furthermore, there are uncertainties, necessitating the management and coordinated scheduling of multiple vehicles. In particular, autonomous specialized work vehicles, such as unmanned express delivery vehicles, unmanned food delivery vehicles, unmanned patrol vehicles, unmanned retail vehicles, and unmanned street sweepers, require effective adjustments and improvements to their autonomous driving technology. This requires unified management and scheduling of all autonomous specialized work vehicles through a cloud-based control platform, effectively controlling their autonomous driving and ensuring their stability, safety, and operational efficiency. The interaction between vehicles and cloud-based control platforms is a core technology crucial for the large-scale commercialization of autonomous specialized work vehicles. However, current interaction technologies between autonomous specialized work vehicles and cloud-based control platforms suffer from poor generalizability, resulting in low operational efficiency and insufficient driving stability. Therefore, it is necessary to develop an interaction method for autonomous specialized work vehicles that can enhance the generalizability of their interaction technology, improve operational efficiency, and enhance driving stability.

[0072] To this end, the embodiment of the present application provides an interaction method and system for an autonomous driving dedicated operation vehicle, for realizing the interaction between the autonomous driving controller, the cloud control platform and the operation device controller, see the attached Figure 1 , including the following steps:

[0073] Step A01) The autonomous driving controller receives the vehicle registration signal and map information, performs path planning based on the vehicle registration signal and map information, and sends the path planning to the cloud control platform.

[0074] Step A02) The autonomous driving controller controls the vehicle to run to the designated location according to the path planning, and determines in real time whether it encounters an obstacle. If an obstacle is encountered, the vehicle driving mode is switched to the remote takeover mode. If no obstacle is encountered, the vehicle is controlled to run to the designated location in the autonomous driving mode.

[0075] Step A03) When the vehicle arrives at the designated location, the cloud control platform sends an operation instruction to the operation device controller.

[0076] Among them, the cloud control platform receives the request information of the user client, forms an operation instruction according to the request information, and sends the operation instruction to the operation device controller.

[0077] Step A04) The operating device controller executes the operating instructions and identifies the completion status of the operation, and sends the completion status of the operation to the cloud control platform.

[0078] By clearly describing the interaction process and data content between dedicated autonomous vehicles and the cloud-based control platform, the technology for their interaction has been standardized. Furthermore, through path planning and the ability to switch driving modes when encountering obstacles, the vehicle's driving stability and operational efficiency have been improved, enabling the complete implementation of autonomous driving tasks, vehicle control and scheduling, and remote takeover of operations when encountering obstacles.

[0079] On the other hand, in this embodiment, the vehicle registration signal includes the vehicle ID and vehicle location information, and the map information includes the map of the area where the vehicle is currently located and the designated location information. Figure 2 , the path planning methods include:

[0080] Step B01) obtaining all candidate navigation paths from the vehicle position to the designated location based on the vehicle position information, a map of the area where the vehicle is currently located, and the designated location information;

[0081] Step B02) extracting static data of sections of candidate navigation paths;

[0082] Step B03) reading historical road segment dynamic data of candidate navigation routes;

[0083] Step B04) Calculate the cost values ​​of all candidate navigation paths based on the static data of the road segment and the dynamic data of the historical road segment, and use the candidate navigation path with the smallest cost value as the vehicle navigation path for this operation.

[0084] The autonomous driving controller integrates, analyzes, and processes the obtained traffic condition data and driving route data based on the vehicle's location information, the map of the vehicle's current area, and the specified location information, while taking into account multiple sources of information such as static and dynamic road section data. It calculates the cost of all candidate navigation paths, selects the candidate navigation path with the smallest cost as the vehicle navigation path for this operation, and then plans an efficient and reasonable navigation path for the autonomous driving dedicated operation vehicle.

[0085] On the other hand, in this embodiment, the method for extracting static data of road segments of candidate navigation paths includes:

[0086] Loading a map of the area where the autonomous driving dedicated operation vehicle is currently located;

[0087] All candidate navigation paths are segmented and lane information for each segment is extracted. Lane information includes segment length l, segment width W, segment flatness P, number of intersections I, and road curvature C. The length L and number of segments N of all candidate navigation paths are obtained.

[0088] Lane information, the length of the candidate navigation path, and the number of road segments constitute the road segment static data.

[0089] On the other hand, in this embodiment, the length L of the candidate navigation path is the sum of the lengths l of each road segment, and the number of road segments in the candidate navigation path N = L ÷ L r , where L r Divide the resolution into preset sections.

[0090] On the other hand, in this embodiment, the historical road segment dynamic data includes the obstacle flow F, congestion probability G, and congestion degree E of the road segment. The cost value of the candidate navigation path is calculated by:

[0091] The length cost of the path: L cost =L×S L ,

[0092] Intersection quantity cost: I cost =I×S I ,

[0093] The width cost of the path: W cost =S W ÷(W1+W2+…+W N ),

[0094] The flatness cost of the path: P cost =S P ÷(P1+P2+…+P N ),

[0095] The path curvature cost value: Ccost =(C1+C2+…+C N )×S C ,

[0096] Obstacle flow cost of the path: F cost =(F1+F2+…+F N )×D F ,

[0097] Congestion probability and degree cost of the path: J cost =(G1×E1+G2×E2+…+G N ×E N )×D J ,

[0098] The total cost of the candidate navigation path is:

[0099] T cost =L cost +I cost +W cost +P cost +C cost +F cost +J cost

[0100] Among them, S L is the length cost coefficient of the preset path, S I is the preset intersection number cost coefficient, S W is the width cost coefficient of the preset path, W1, W2, ... W N is the width of all road segments in the path, S P is the flatness cost coefficient of the preset path, P1, P2, ...P N is the flatness of all sections in the path, C1, C2, ...C N is the road curvature of all road segments in the path, S C are the road curvature cost coefficients of the preset path, F1, F2, ...F N is the obstacle flow of all sections in the path, D F are the obstacle flow cost coefficients of the preset path, G1, G2, ... G N is the congestion probability of all road sections in the path, E1, E2, ...E N is the congestion level of all road sections in the path, D J It is the congestion probability and degree cost coefficient of the preset path.

[0101] On the other hand, in this embodiment, the obstacle flow F includes motor vehicle flow V, non-motor vehicle flow B, pedestrian flow X, and unknown obstacle flow U. The method for calculating the obstacle flow cost value of the candidate navigation path also includes:

[0102] The cost of vehicle traffic flow on the path: V cost =(V1+V2+…+V N )×D V ,

[0103] Non-motorized vehicle flow cost value of the path: B cost =(B1+B2+…+B N )×D B ,

[0104] Pedestrian flow cost value of the path: X cost =(X1+X2+…+X N )×D X ,

[0105] The unknown obstacle flow cost of the path: U cost =(U1+U2+…+U N )×D U ,

[0106] The obstacle flow cost of the candidate navigation path is:

[0107] F cost =V cost +B cost +X cost +U cost

[0108] Among them, V1, V2, ... V N is the motor vehicle flow of all road sections in the route, D V are the motor vehicle flow cost coefficients of the preset route, B1, B2, ...B N is the non-motor vehicle flow rate of all road sections in the route, D B is the non-motor vehicle flow cost coefficient of the preset path, X1, X2, ...X N is the pedestrian flow of all road sections in the path, D X is the pedestrian flow cost coefficient of the preset path, U1, U2, ...U N is the unknown obstacle flow of all sections in the path, D U is the flow cost coefficient of the unknown obstacle in the preset path.

[0109] On the other hand, in this embodiment, the method for determining whether an obstacle is encountered includes:

[0110] The autonomous driving controller updates the vehicle’s positioning information in real time during operation;

[0111] If the positioning information does not change after exceeding the preset time threshold, it is determined that the vehicle has encountered an obstacle. If the positioning information changes in real time, it is determined that the vehicle has not encountered an obstacle.

[0112] Optionally, the above method for determining whether an obstacle is encountered can also be used to determine whether the vehicle is stuck in a pothole on the road, whether the vehicle has triggered the emergency braking function, whether the positioning sensor of the automatic driving controller has failed, etc.

[0113] On the other hand, in this embodiment, the method for switching the vehicle driving mode to the remote takeover mode after determining that the vehicle encounters an obstacle includes:

[0114] When the autonomous driving controller determines that the vehicle encounters an obstacle and the autonomous driving mode cannot bypass it, it reports to the cloud control platform and switches the vehicle's driving mode to remote takeover mode;

[0115] The automatic driving controller receives and executes the vehicle movement remote control instructions sent by the remote duty personnel to the vehicle until the vehicle bypasses the obstacle.

[0116] Taking unmanned delivery vehicles suitable for use in communities, office parks, and other scenarios as an example, the implementation process of the embodiment of this application is as follows:

[0117] The autonomous driving controller receives the registration signal and map information of the unmanned delivery vehicle. Figure 3 , based on the registration signal and the location information of the unmanned delivery vehicle in the map information, the map of the specific community, office park or park, scenic area where the unmanned delivery vehicle is currently located, and the delivery point information, all candidate navigation paths from the location of the unmanned delivery vehicle to the delivery point are obtained;

[0118] Extract the static data of the candidate navigation path; read the dynamic data of the historical sections of the candidate navigation path; calculate the cost values ​​of all candidate navigation paths based on the static data of the sections and the dynamic data of the historical sections, and use the candidate navigation path with the smallest cost value as the navigation path for the unmanned delivery vehicle running this time. Figure 4 And send the navigation path of the unmanned delivery vehicle to the cloud control platform;

[0119] By default, the vehicle is in a standby state, meaning it is parked at an initial fixed delivery point. The courier will deliver the package to the unmanned delivery vehicle. The recipient can use the client to select a delivery location and click Deliver, which will deliver the package to the designated location C selected by the user.

[0120] The autonomous driving controller controls the unmanned delivery vehicle to the designated location C according to the navigation path of the unmanned delivery vehicle, and determines in real time whether it encounters an obstacle. If an obstacle is encountered, the unmanned delivery vehicle driving mode is switched to remote takeover mode. If no obstacle is encountered, the unmanned delivery vehicle is controlled to run to the designated location C in the autonomous driving mode.

[0121] When the unmanned delivery vehicle arrives at the designated location C, the cloud control platform sends an operation waiting instruction to the operation device controller and automatically notifies the recipient to pick up the package;

[0122] The operating device controller executes the operation waiting instruction and keeps the vehicle parked and waiting. The recipient clicks the screen of the express cabinet to start the pickup operation. The cloud control platform sends the operation in progress instruction to the operating device controller. The unmanned express vehicle keeps parked and waiting. After the recipient picks up the package and closes the cabinet door, the cloud control platform sends the operation completion status to the cloud control platform, which then sends the operation completion status to the client.

[0123] Please see the attached Figure 5 When the recipient at the designated location C completes the operation or the waiting time expires, the unmanned delivery vehicle automatically moves to the next designated location E. The recipient can check the delivery status through mobile apps, mini-programs, and other clients.

[0124] Please see the attached Figure 6 ,When the recipient at the designated location E completes the pickup or the waiting time has timed out, and no other user requests express delivery, the vehicle automatically returns to the initial express delivery point;

[0125] When the unmanned delivery vehicle waits for the recipient to pick up the package for longer than the preset time threshold, the cloud control platform sends an unfinished operation status to the operation device controller, and at the same time reminds the recipient that the waiting time for pickup has timed out and needs to be delivered again.

[0126] The embodiments of this application clearly and explicitly illustrate the interaction process and data content between the dedicated autonomous driving work vehicle and the cloud control platform, improving the universality of the interaction technology between the dedicated autonomous driving work vehicle and the cloud control platform. The relevant interactive data content or interaction steps during the interaction process between the dedicated autonomous driving work vehicle and the cloud control platform can be added, deleted, or adjusted according to different vehicle models or different work tasks (such as express delivery, food delivery, sales, patrol, sweeping, etc.). At the same time, the embodiments of this application improve the driving stability and operating efficiency of the dedicated autonomous driving work vehicle, thereby better realizing functions such as complete autonomous driving work tasks, vehicle control and scheduling, and remote takeover mode operation when encountering obstacles.

[0127] On the other hand, in this embodiment, the cloud control platform records the detour starting point position and detour path in the remote takeover mode;

[0128] Segment the multiple detour paths recorded at the same detour starting point, making each segment a straight line segment or a circular arc segment, record the direction and length of the straight line segment, and record the radius and radian of the circular arc segment;

[0129] Set a similarity threshold, and consider two straight line segments whose similarity exceeds the similarity threshold as similar segments, and consider two arc segments whose similarity exceeds the similarity threshold as similar segments;

[0130] Counting the number of similar segments of each straight line segment and arc segment, and recording the straight line segment and arc segment whose number of similar segments exceeds a preset threshold as the preferred detour operation;

[0131] When the subsequent vehicle runs to the detour starting point again, the automatic driving controller determines whether the obstacle still exists. If the obstacle still exists, the automatic driving controller attempts to detour according to the preferred detour operation. If the obstacle disappears, the vehicle runs to the designated position in automatic driving mode.

[0132] The operations for calculating the similarity of the straight line segments and the arc segments are similar to the operations for calculating the similarity of the classification algorithm in the prior art, and will not be described in detail in the embodiments of the present application.

[0133] The cloud control platform is used to perform segmentation analysis and similarity calculation on the detour starting point position and detour path recorded in the remote takeover mode, so as to plan a more efficient and reasonable detour route for the autonomous driving dedicated work vehicle when it encounters obstacles.

[0134] On the other hand, in this embodiment, the method of attempting to bypass includes:

[0135] Execute any preferred detour operation until there are no obstacles when re-routing or the preferred detour operation is executed up to the preset upper limit. If the preferred detour operation is executed up to the preset upper limit, report to the cloud control platform and switch the vehicle driving mode to remote takeover mode.

[0136] By presetting an upper limit on the number of times the preferred detour operation can be performed, and reporting to the cloud control platform to switch the vehicle driving mode when the preset upper limit is exceeded, the obstacle detour route can be adjusted in time, and the operation task of the dedicated autonomous driving operation vehicle can be successfully completed.

[0137] On the other hand, the embodiment of the present application also provides an interactive system for an autonomous driving dedicated operation vehicle, see the attached Figure 7 ,include:

[0138] The automatic driving controller 1 is used to receive the vehicle registration signal and map information, perform path planning based on the vehicle registration signal and map information, control the vehicle to a designated location based on the path planning, and determine in real time whether an obstacle is encountered. If an obstacle is encountered, the vehicle driving mode is switched to a remote takeover mode. If no obstacle is encountered, the vehicle is controlled to drive to the designated location in the automatic driving mode.

[0139] Cloud control platform 2, used to send operation instructions to the operation device controller when the vehicle arrives at the designated location;

[0140] The operation device controller 3 is used to execute the operation instructions and identify the completion status of the operation;

[0141] The vehicle-mounted T-BOX4 is used to realize communication and transmission between the automatic driving controller 1, the working device controller 3 and the cloud control platform 2.

[0142] Among them, the on-board T-BOX4 (Telematics BOX) is a remote information processor used to realize the transmission of instructions and information. The on-board T-BOX4 can deeply read the car CAN bus data and private protocols. Among them, CAN (Controller Area Network) is an ISO internationally standardized serial communication protocol, and the CAN bus is one of the most widely used field buses in the world. Generally speaking, when the user sends a control command through the mobile phone APP, the background will send a monitoring request instruction to the on-board T-BOX4. After the vehicle obtains the control command, it sends a control message through the CAN bus and realizes the control of the vehicle. Finally, the operation results are fed back to the user's mobile phone APP. This function alone can help users remotely start the vehicle, turn on the air conditioner, adjust the seat to the appropriate position, etc.

[0143] In this embodiment, the communication between the autonomous driving controller 1 and the working device controller 3 and the vehicle-mounted T-BOX 4 is carried out via CAN or vehicle Ethernet, and the communication between the cloud control platform 2 and the vehicle-mounted T-BOX 4 is carried out via 4G or 5G. These communications primarily involve sending the path plan generated by the autonomous driving controller 1 to the cloud control platform 2, sending the work instructions generated by the cloud control platform 2 to the working device controller 3, and sending the work completion status identified by the working device controller 3 to the cloud control platform 2.

[0144] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Those skilled in the art will understand that the present invention includes, but is not limited to, the contents described in the drawings and the above specific embodiments. Any modifications that do not deviate from the functional and structural principles of the present invention are intended to be included within the scope of the claims.

Claims

1. An interaction method for an autonomous driving dedicated work vehicle, used to achieve interaction between an autonomous driving controller, a cloud control platform, and a work device controller, characterized in that: The following steps are involved: The autonomous driving controller receives a vehicle registration signal and map information, wherein the vehicle registration signal includes a vehicle ID and vehicle location information, and the map information includes a map of an area where the vehicle is currently located and designated location information; Obtain all candidate navigation paths from the vehicle position to the designated location based on the vehicle position information, a map of the area the vehicle is currently located in, and the designated location information; Extracting static data of the sections of the candidate navigation paths, wherein the static data includes the number of intersections I, the width of the section W, the flatness of the section P, the curvature of the road C, the length L of all candidate navigation paths, and the number N of sections; Reading historical road segment dynamic data of the candidate navigation path, wherein the historical road segment dynamic data includes obstacle flow F, congestion probability G, and congestion degree E of the road segment; Calculate the cost of all candidate navigation paths based on the static data of the road section and the dynamic data of the historical road section, use the candidate navigation path with the smallest cost as the vehicle navigation path for this operation, and send the path planning to the cloud control platform; The automatic driving controller controls the vehicle to run to the designated location according to the path planning, and determines in real time whether it encounters an obstacle. If an obstacle is encountered, the vehicle driving mode is switched to the remote takeover mode. If no obstacle is encountered, the vehicle is controlled to run to the designated location in the automatic driving mode. When the vehicle arrives at the designated location, the cloud control platform sends an operation instruction to the operation device controller; The operation device controller executes the operation instruction and identifies the operation completion status, and sends the operation completion status to the cloud control platform; The method for calculating the cost value of the candidate navigation path includes: The length of the path cost: , The cost of the number of intersections: , The width cost of the path: , The flatness cost of the path: , The road curvature cost of the path: , Obstacle flow cost of the path: , Congestion probability and degree cost of the path: , The total cost of the candidate navigation path is: ， in, is the length cost coefficient of the preset path, is the preset intersection number substitution coefficient, is the width cost coefficient of the preset path, is the width of all road segments in the path, is the flatness cost coefficient of the preset path, is the flatness of all sections in the path, is the road curvature of all road segments in the path, is the road curvature cost coefficient of the preset path, is the obstacle flow of all sections in the path, is the obstacle flow cost coefficient of the preset path, is the congestion probability of all road segments in the path, is the congestion level of all road segments in the route, It is the congestion probability and degree cost coefficient of the preset path.

2. The interactive method for an autonomous driving dedicated work vehicle according to claim 1, characterized in that: The method for extracting static data of the road section of the candidate navigation path includes: Loading a map of the area where the autonomous driving dedicated operation vehicle is currently located; All candidate navigation paths are segmented and lane information for each segment is extracted. The lane information includes segment length l, segment width W, segment flatness P, number of intersections I, and road curvature C. The length L and number of segments N of all candidate navigation paths are obtained. The lane information, the length of the candidate navigation path, and the number of road sections constitute road section static data.

3. The interactive method for an autonomous driving dedicated work vehicle according to claim 1, characterized in that: The obstacle flow F includes motor vehicle flow V, non-motor vehicle flow B, pedestrian flow X, and unknown obstacle flow U. The method for calculating the obstacle flow cost value of the candidate navigation path also includes: The cost of vehicle traffic flow on the path: , The cost of non-motorized vehicle traffic flow on the path: , Pedestrian flow cost of the path: , The unknown obstacle flow cost of the path: , The obstacle flow cost of the candidate navigation path is: ， in, is the motor vehicle flow cost coefficient of the preset path, is the non-motor vehicle flow cost coefficient of the preset path, is the pedestrian flow cost coefficient of the preset path, is the flow cost coefficient of the unknown obstacle in the preset path.

4. The interactive method for an autonomous driving dedicated work vehicle according to claim 1, characterized in that: The method for determining whether an obstacle is encountered includes: The autonomous driving controller updates the vehicle’s positioning information in real time during operation; If the positioning information does not change after exceeding a preset time threshold, it is determined that the vehicle has encountered an obstacle. If the positioning information changes in real time, it is determined that the vehicle has not encountered an obstacle.

5. The interactive method for an autonomous driving dedicated work vehicle according to claim 4, characterized in that: The method for switching the vehicle driving mode to the remote takeover mode after determining that the vehicle encounters an obstacle includes: When the autonomous driving controller determines that the vehicle encounters an obstacle and the autonomous driving mode cannot bypass it, it reports to the cloud control platform and switches the vehicle driving mode to the remote takeover mode; The automatic driving controller receives and executes the vehicle movement remote control instruction sent by the remote duty personnel to the vehicle until the vehicle bypasses the obstacle.

6. The interactive method for an autonomous driving dedicated work vehicle according to claim 5, characterized in that: The cloud control platform records the detour starting point and detour path in the remote takeover mode; Segmenting the multiple detour paths recorded at the same detour starting point, making each segment a straight line segment or a circular arc segment, recording the direction and length of the straight line segment, and recording the radius and radian of the circular arc segment; Setting a similarity threshold, treating two straight line segments whose similarity exceeds the similarity threshold as similar segments, and treating two arc segments whose similarity exceeds the similarity threshold as similar segments; Counting the number of similar segments of each straight line segment and circular arc segment, and recording the straight line segment and circular arc segment whose number of similar segments exceeds a preset threshold as the preferred detour operation; When the subsequent vehicle runs to the detour starting point again, the automatic driving controller determines whether the obstacle still exists. If the obstacle still exists, the automatic driving controller attempts to detour according to the preferred detour operation. If the obstacle disappears, the vehicle runs to the designated position in the automatic driving mode.

7. The interactive method for an autonomous driving dedicated work vehicle according to claim 6, characterized in that: The methods of attempting to bypass include: Execute any of the preferred detour operations until there are no obstacles when re-planning the path or the preferred detour operation is executed up to a preset upper limit. If the preferred detour operation is executed up to a preset upper limit, report to the cloud control platform and switch the vehicle driving mode to the remote takeover mode.

8. An interactive system for an autonomous driving dedicated work vehicle, characterized in that: include: an autonomous driving controller, configured to receive a vehicle registration signal and map information, the vehicle registration signal including a vehicle ID and vehicle location information, the map information including a map of the vehicle's current area and designated location information, perform path planning based on the vehicle registration signal and map information, control the vehicle to a designated location based on the path planning, and determine in real time whether an obstacle is encountered. If an obstacle is encountered, the vehicle driving mode is switched to a remote takeover mode; if no obstacle is encountered, the vehicle is controlled to operate in autonomous driving mode to the designated location; The cloud control platform is used to send operation instructions to the operation device controller when the vehicle reaches the designated location; An operating device controller, configured to execute the operating instructions and identify the completion status of the operating; The vehicle-mounted T-BOX is used to achieve communication and transmission between the autonomous driving controller, the operating device controller, and the cloud control platform; The autopilot controller is also used to perform the following steps: Obtain all candidate navigation paths from the vehicle position to the designated location based on the vehicle position information, a map of the area the vehicle is currently located in, and the designated location information; Extracting static data of the sections of the candidate navigation paths, wherein the static data includes the number of intersections I, the width of the section W, the flatness of the section P, the curvature of the road C, the length L of all candidate navigation paths, and the number N of sections; Reading historical road segment dynamic data of the candidate navigation path, wherein the historical road segment dynamic data includes obstacle flow F, congestion probability G, and congestion degree E of the road segment; Based on the static data of the road section and the dynamic data of the historical road section, the cost values ​​of all candidate navigation paths are calculated, the candidate navigation path with the smallest cost value is used as the vehicle navigation path for this operation, and the path planning is sent to the cloud control platform, wherein the cost value calculation method of the candidate navigation path includes: The length of the path cost: , The cost of the number of intersections: , The width cost of the path: , The flatness cost of the path: , The road curvature cost of the path: , Obstacle flow cost of the path: , Congestion probability and degree cost of the path: , The total cost of the candidate navigation path is: ， in, is the length cost coefficient of the preset path, is the preset intersection number substitution coefficient, is the width cost coefficient of the preset path, is the width of all road segments in the path, is the flatness cost coefficient of the preset path, is the flatness of all sections in the path, is the road curvature of all road segments in the path, is the road curvature cost coefficient of the preset path, is the obstacle flow of all sections in the path, is the obstacle flow cost coefficient of the preset path, is the congestion probability of all road segments in the path, is the congestion level of all road segments in the route, It is the congestion probability and degree cost coefficient of the preset path.

Images