Multi-vehicle global path planning method

By identifying and integrating the unmanned vehicle path map, calculating priority values, and optimizing path planning, the problem of low transportation efficiency of multiple vehicles in industrial parks is solved, global path planning is realized, and transportation efficiency and adaptability are improved.

CN118913301APending Publication Date: 2024-11-08ZHONGWEI WUCHUANG INTELLIGENT TECH (SHANGHAI) CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410968316.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-18
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

The existing single path planning method is difficult to meet the transportation needs of multiple vehicles in complex industrial parks, and cannot effectively improve overall transportation efficiency and meet the use needs of enterprises.

Method used

By identifying the path map of unmanned vehicles in the working area, establishing an initial path model, integrating dynamic analysis units and positioning adjustment units, calculating the priority values ​​of each combination method, performing global path planning, optimizing vehicle paths and driving solutions, and realizing multi-vehicle Comprehensive analysis and navigation path planning.

Benefits of technology

It improves overall transportation efficiency, can correct the item adjustment coefficient according to the progress plan, achieve global adjustment, meet the planning needs of multiple unmanned vehicle paths, and adapt to complex road conditions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118913301A_ABST
    Figure CN118913301A_ABST
Patent Text Reader

Abstract

The invention discloses a multi-vehicle global path planning method, and belongs to the technical field of unmanned driving, and the method specifically comprises the steps: 1, recognizing a working region, and drawing an unmanned vehicle path diagram in the working region; 2, establishing a corresponding path initial model based on the unmanned vehicle path diagram, and integrating the path initial model to obtain a monitoring model; 3, identifying the combination modes of the to-be-selected paths in the monitoring model, and calculating the priority values of the combination modes; 4, performing global path planning according to the calculated priority value; through comprehensive analysis of unmanned vehicles in all working areas, an adjustment coefficient which can improve the overall transportation efficiency as much as possible and can manually correct each article according to a progress plan is planned, so that global adjustment is realized, and global planning of paths of multiple unmanned vehicles is realized. And the use requirements of enterprises can be better met.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the technical field of unmanned driving, and specifically relates to a multi-vehicle global path planning method. Background Art

[0002] In recent years, the development of artificial intelligence technology has brought revolutionary changes to the unmanned driving industry. The current implementation of path planning involves multiple disciplines and fields, including computer software, algorithms, data structures, industrial control hardware, object recognition, sensor perception, etc. It uses sensors with different functions to perceive the vehicle's surrounding environment, and plans a safe and collision-free path based on the road, vehicle position and obstacle information obtained by perception, controls the vehicle's speed and steering, so that the vehicle can safely drive autonomously on the road. At the same time, unmanned driving has also been widely used in the field of industrial parks;

[0003] However, as more and more unmanned vehicles are used to transport goods in industrial parks, the road conditions used by unmanned vehicles are more complicated. Currently, a single path planning can no longer meet the usage requirements. Therefore, the present invention provides a multi-vehicle global path planning method. Summary of the invention

[0004] In order to solve the problems existing in the above scheme, the present invention provides a multi-vehicle global path planning method.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] A multi-vehicle global path planning method, the specific method comprising:

[0007] Step 1: Identify the work area and draw a path map of the unmanned vehicle in the work area;

[0008] Step 2: Establish a corresponding path initial model based on the unmanned vehicle path map, integrate the path initial model, and obtain a monitoring model;

[0009] Step 3: Identify the combination of each candidate path in the monitoring model and calculate the priority value of each combination;

[0010] Step 4: Perform global path planning based on the calculated priority value.

[0011] Furthermore, the method of integrating the path initial model includes:

[0012] An unmanned vehicle model library is established to obtain the type of unmanned vehicles working in the working area, and the corresponding unmanned vehicle model is matched from the unmanned vehicle model library according to the obtained unmanned vehicle type, and the obtained unmanned vehicle model is added to the path initial model. A positioning adjustment unit and a dynamic analysis unit are set in the path initial model. The positioning adjustment unit is used to receive the positioning information of each unmanned vehicle in real time, and update the position of the corresponding unmanned vehicle model in the path initial model in real time according to the received positioning information; the dynamic analysis unit is used to dynamically display the time when the corresponding unmanned vehicle arrives at the position on each path in the path initial model according to the current information of the unmanned vehicle, and mark the current path initial model as a monitoring model.

[0013] Furthermore, the working method of the dynamic analysis unit includes:

[0014] The speed, current position and target position of the unmanned vehicle are obtained in real time, the candidate paths for the unmanned vehicle to reach the target position from the current position are identified, the arrival time of the unmanned vehicle at each position on each candidate path is calculated according to the standard time formula, the arrival time of each unmanned vehicle at each position on each candidate path is identified, the corresponding conflict points are marked on the candidate paths according to the driving direction of each unmanned vehicle, the transfer point closest to the conflict point is identified, the conflict selection is performed, the corresponding update is performed according to the conflict selection, and so on, until no conflict points appear.

[0015] Furthermore, the method for performing conflict selection includes:

[0016] Identify the change value of the conflicting unmanned vehicles, mark the obtained change value as BD, identify the type of items transported by each conflicting unmanned vehicle, match the corresponding adjustment coefficient according to the identified type of items, mark it as α, calculate the corresponding conflict value according to the formula QX=BD×α, and select the change method corresponding to the maximum conflict value for change.

[0017] Furthermore, the method for performing conflict selection includes:

[0018] Identify the conflicting vehicles that arrive at the transfer point first, select the conflicting vehicles that arrive at the transfer point first to wait, mark them as waiting vehicles, and calculate the waiting time of the waiting vehicles.

[0019] Furthermore, the method for calculating the priority value of each combination includes:

[0020] Mark each unmanned vehicle in the combination as i, where i = 1, 2, ..., n, and n is a positive integer; identify the change duration corresponding to each unmanned vehicle in the combination, mark the obtained change duration as BHi, match the corresponding adjustment coefficient αi according to the type of goods transported by each unmanned vehicle, and use the formula Calculate the corresponding priority value.

[0021] Furthermore, the method for performing global path planning according to the calculated priority value includes:

[0022] Identify the corresponding combination with the highest priority value, mark it as a planning combination, identify the candidate paths corresponding to each unmanned vehicle in the planning combination, use the identified candidate paths as navigation paths for each unmanned vehicle, and send the corresponding navigation paths and driving plans to the corresponding unmanned vehicles.

[0023] Compared with the prior art, the present invention has the following beneficial effects:

[0024] By conducting a comprehensive analysis of the unmanned vehicles in all work areas, a plan is made to maximize the overall transportation efficiency. The adjustment coefficients of each item can be manually corrected according to the schedule, thereby achieving global adjustment and realizing global planning of multiple unmanned vehicle routes, which can better meet the needs of enterprises. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0026] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0027] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. 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.

[0028] like Figure 1 As shown, a multi-vehicle global path planning method, the specific method includes:

[0029] Step 1: Identify the work area and draw a path map of the unmanned vehicle in the work area;

[0030] The working area is the area where the unmanned transport vehicle needs to work; the unmanned vehicle is the unmanned transport vehicle; the unmanned vehicle path map is the distribution map of feasible paths in the working area, which can be drawn accordingly through existing methods.

[0031] Step 2: Establish a corresponding path initial model based on the unmanned vehicle path map, integrate the path initial model, and obtain a monitoring model;

[0032] The path initial model is to establish a corresponding three-dimensional data model based on the existing three-dimensional modeling technology.

[0033] Methods for integrating the path initialization model include:

[0034] An unmanned vehicle model library is established to obtain the type of unmanned vehicles working in the working area, and the corresponding unmanned vehicle model is matched from the unmanned vehicle model library according to the obtained unmanned vehicle type, and the obtained unmanned vehicle model is added to the path initial model. A positioning adjustment unit and a dynamic analysis unit are set in the path initial model. The positioning adjustment unit is used to receive the positioning information of each unmanned vehicle in real time, and update the position of the corresponding unmanned vehicle model in the path initial model in real time according to the received positioning information; the dynamic analysis unit is used to dynamically display the time when the corresponding unmanned vehicle arrives at the position on each path in the path initial model according to the current information of the unmanned vehicle, and mark the current path initial model as a monitoring model.

[0035] The application functions of the positioning adjustment unit can be implemented accordingly through the existing technology, so it will not be described in detail.

[0036] The working methods of the dynamic analysis unit include:

[0037] The speed, current position and target position of the unmanned vehicle are obtained in real time, the candidate paths for the unmanned vehicle to reach the target position from the current position are identified, and the arrival time of the unmanned vehicle at each position on each candidate path is calculated according to the standard time formula. The standard time formula is based on the path length divided by the current speed and then added to the current time, that is, the standard time formula is calculated without other influences; the arrival time of each unmanned vehicle at each position on each candidate path is identified, and the corresponding conflict points are marked on the candidate paths according to the driving direction of each unmanned vehicle, the transfer point closest to the conflict point is identified, and the conflict selection is performed. According to the conflict selection, corresponding updates are performed, and so on, until no conflict points appear.

[0038] Conflict points are points that are blocked due to driving direction, channel width, etc., and cannot be passed. Because in the management of industrial parks, in order to improve the space utilization rate of industrial parks, internal channels are often only one-way streets, that is, only one unmanned vehicle loaded with goods is allowed to pass in the same direction.

[0039] Identify the transfer point closest to the conflict point. The closest refers to the closest point relative to the conflicting vehicle, taking into account turning back, etc. If it is not on the driving path, the corresponding turning back distance needs to be considered. The corresponding nearest transfer point can be identified through the existing method. The transfer point refers to several areas set up in the working area for vehicle staggering, that is, multiple unmanned vehicles can be allowed to pass through at the same time at the transfer point.

[0040] In one embodiment, a method for performing conflict selection includes:

[0041] Identify the change value of the conflicting unmanned vehicles, mark the obtained change value as BD, identify the type of items transported by each conflicting unmanned vehicle, match the corresponding adjustment coefficient according to the identified type of items, mark it as α, calculate the corresponding conflict value according to the formula QX=BD×α, and select the change method corresponding to the maximum conflict value for change.

[0042] The change value refers to the convenience value of the change method adopted by each conflicting vehicle if it is desired to allow the conflicting vehicles to pass through at the nearest transfer point in a staggered manner. For example, the conflicting vehicle that arrives later can increase its speed, and the one that arrives earlier can reduce its speed or wait after arriving first, etc., to evaluate the impact of possible methods, such as beneficial impact, improving transportation efficiency, adverse impact, causing more vehicle conflicts, speed change, and potential safety hazards. Specifically, a corresponding conflict analysis model is established based on a CNN network or a DNN network, and the corresponding training set is set manually for training. The conflict analysis model after successful training is used for analysis to obtain the corresponding change value. Because neural networks are existing technologies in this field, the specific establishment and training process will not be described in detail. Generally, the first-arrival waiting method is adopted, and the original waiting time at a constant speed is multiplied by a preset coefficient to obtain the change value, and the conflict analysis model mainly analyzes the preset coefficients corresponding to each method and then multiplies them.

[0043] The corresponding adjustment coefficient is matched according to the identified item type. The corresponding adjustment coefficient is set manually according to the item type, and a corresponding adjustment coefficient matching table is established, and then the matching is performed and the result is obtained.

[0044] In one embodiment, a method for performing conflict selection includes:

[0045] Identify the conflicting vehicles that arrive at the transfer point first, select the conflicting vehicles that arrive at the transfer point first to wait, mark them as waiting vehicles, and calculate the waiting time of the waiting vehicles.

[0046] Step 3: Identify the combination of each candidate path in the monitoring model and calculate the priority value of each combination;

[0047] Identify the combination of each candidate path in the monitoring model, that is, obtain the combination according to the candidate path combination that the unmanned vehicle may have. The monitoring model has different arrival times of different candidate paths, that is, the corresponding combination can be identified through the existing method.

[0048] Methods for calculating the priority values ​​of various combinations include:

[0049] Mark each unmanned vehicle in the combination as i, where i = 1, 2, ..., n, and n is a positive integer; identify the change duration corresponding to each unmanned vehicle in the combination, mark the obtained change duration as BHi, match the corresponding adjustment coefficient αi according to the type of goods transported by each unmanned vehicle, which is the same as that in the calculation of the conflict value; according to the formula Calculate the corresponding priority value.

[0050] The change duration is calculated based on a certain combination, that is, based on one combination, the change duration of each unmanned vehicle reaching the target position in other combinations is calculated. The shortened time is positive, otherwise it is negative.

[0051] Step 4: Perform global path planning based on the calculated priority value.

[0052] Methods for global path planning based on calculated priority values ​​include:

[0053] Identify the corresponding combination with the highest priority value, mark it as a planning combination, identify the candidate paths corresponding to each unmanned vehicle in the planning combination, use the identified candidate paths as navigation paths for each unmanned vehicle, and send the corresponding navigation paths and driving plans to the corresponding unmanned vehicles.

[0054] The driving plan includes speed changes, stops at transfer points, etc.

[0055] The above formulas are all calculated by removing dimensions and taking numerical values. The formula is a formula that is closest to the actual situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained by simulating a large amount of data.

[0056] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A multi-vehicle global path planning method, characterized in that: Specific methods include: Step 1: Identify the work area and draw a path map of the unmanned vehicle in the work area; Step 2: Establish a corresponding path initial model based on the unmanned vehicle path map, integrate the path initial model, and obtain a monitoring model; Step 3: Identify the combination of each candidate path in the monitoring model and calculate the priority value of each combination; Step 4: Perform global path planning based on the calculated priority value.

2. A multi-vehicle global path planning method according to claim 1, characterized in that: Methods for integrating the path initialization model include: An unmanned vehicle model library is established to obtain the type of unmanned vehicles working in the working area, and the corresponding unmanned vehicle model is matched from the unmanned vehicle model library according to the obtained unmanned vehicle type, and the obtained unmanned vehicle model is added to the path initial model. A positioning adjustment unit and a dynamic analysis unit are set in the path initial model. The positioning adjustment unit is used to receive the positioning information of each unmanned vehicle in real time, and update the position of the corresponding unmanned vehicle model in the path initial model in real time according to the received positioning information; the dynamic analysis unit is used to dynamically display the time when the corresponding unmanned vehicle arrives at the position on each path in the path initial model according to the current information of the unmanned vehicle, and mark the current path initial model as a monitoring model.

3. A multi-vehicle global path planning method according to claim 2, characterized in that: The working methods of the dynamic analysis unit include: The speed, current position and target position of the unmanned vehicle are obtained in real time, the candidate paths for the unmanned vehicle to reach the target position from the current position are identified, the arrival time of the unmanned vehicle at each position on each candidate path is calculated according to the standard time formula, the arrival time of each unmanned vehicle at each position on each candidate path is identified, the corresponding conflict points are marked on the candidate paths according to the driving direction of each unmanned vehicle, the transfer point closest to the conflict point is identified, the conflict selection is performed, the corresponding update is performed according to the conflict selection, and so on, until no conflict points appear.

4. A multi-vehicle global path planning method according to claim 3, characterized in that: Methods for making conflict selections include: Identify the change value of the conflicting unmanned vehicles, mark the obtained change value as BD, identify the type of items transported by each conflicting unmanned vehicle, match the corresponding adjustment coefficient according to the identified type of items, mark it as α, calculate the corresponding conflict value according to the formula QX=BD×α, and select the change method corresponding to the maximum conflict value for change.

5. A multi-vehicle global path planning method according to claim 3, characterized in that: Methods for making conflict selections include: Identify the conflicting vehicles that arrive at the transfer point first, select the conflicting vehicles that arrive at the transfer point first to wait, mark them as waiting vehicles, and calculate the waiting time of the waiting vehicles.

6. A multi-vehicle global path planning method according to claim 1, characterized in that: Methods for calculating the priority values ​​of various combinations include: Mark each unmanned vehicle in the combination as i, where i = 1, 2, ..., n, and n is a positive integer; identify the change duration corresponding to each unmanned vehicle in the combination, mark the obtained change duration as BHi, match the corresponding adjustment coefficient αi according to the type of goods transported by each unmanned vehicle, and use the formula Calculate the corresponding priority value.

7. A multi-vehicle global path planning method according to claim 6, characterized in that: Methods for global path planning based on calculated priority values ​​include: Identify the corresponding combination with the highest priority value, mark it as a planning combination, identify the candidate paths corresponding to each unmanned vehicle in the planning combination, use the identified candidate paths as navigation paths for each unmanned vehicle, and send the corresponding navigation paths and driving plans to the corresponding unmanned vehicles.