CN-121977596-A - Path planning and operation management system and method for unmanned sweeper

CN121977596ACN 121977596 ACN121977596 ACN 121977596ACN-121977596-A

Abstract

The invention relates to the technical field of intelligent transportation, in particular to a path planning and operation management system and method of an unmanned motor sweeper, which are used for generating and periodically updating a dynamic cleaning value graph reflecting each road section for an operation area by fusing multi-source data, decomposing the area into a plurality of subtask blocks and distributing the subtask blocks to a group of motor sweeper to form a collaborative operation group based on the value graph, a plurality of subtask blocks and a global operation target, planning an initial global path for each group of motor sweeper to avoid space overlapping and operation parameters positively related to cleaning values, monitoring the state of the motor sweeper in real time during operation, triggering task re-planning and re-distribution when dynamic events are carried out, and continuously optimizing a value scoring model and a task distribution strategy by comparing planning and executing data after the tasks are completed. The method solves the problems of low efficiency, repeated coverage or omission and difficulty in coping with dynamic tasks caused by static isolation and lack of cooperation of the conventional motor sweeper path planning based on single-vehicle intelligence.

Inventors

ZHANG ZHONGWEI
YANG JIAMING
SONG TIANCHENG
YUAN XIU
YAN CHUN
ZHOU YUXIN

Assignees

绵阳新投实业有限责任公司

Dates

Publication Date: 20260505
Application Date: 20260119

Claims (10)

1. The path planning and operation management method of the unmanned sweeping vehicle is characterized by comprising the following steps of: acquiring multi-source data from an unmanned sweeper, road side sensing equipment and an external system, generating and periodically updating dynamic value graphs for different road sections in a working area based on the multi-source data, wherein the dynamic value graphs comprise dynamic cleaning value scores of all road sections; Decomposing a target area into a plurality of subtask blocks based on the dynamic value map, the real-time states of all available sweeper vehicles and a preset global operation target, and distributing the subtask blocks to a group of sweeper vehicles to form a cooperative operation group; Generating an initial global path and associated operation control parameters for each sweeper in the collaborative operation group based on the allocated subtask blocks and the dynamic value graphs, wherein the generation of the initial global path is prevented from generating instant spatial overlapping with the current operation areas of other sweeper in the group; During the execution process of the operation, the states of all the sweeper are monitored, and when a preset dynamic event is detected, the task reevaluation and the rescheduling of the affected sweeper are triggered; And after the task is completed, optimizing and iterating the calculation logic of the dynamic cleaning value score and the allocation strategy of the subtask block based on the comparison analysis result of the planning data and the actual execution data.
2. The method for path planning and operation management of an unmanned sweeping vehicle of claim 1, And calculating the dynamic cleaning value score, wherein at least two of historical cleanliness data, a dirt event grade reported in real time, traffic flow data and weather influence factors are integrated.
3. The method for path planning and job management for an unmanned motor sweeper of claim 1, wherein assigning the sub-task blocks to a group of motor sweeper to form a collaborative job group comprises: And calculating based on the real-time position of the vehicle, the remaining operation endurance and the estimated time consumption of driving to each block to achieve the global operation target, wherein the global operation target is to maximize the total finished cleaning value or balance the operation load of each vehicle in a specified time window.
4. The method of path planning and job management for unmanned sweeping vehicles of claim 1, wherein generating an initial global path and associated job control parameters for each sweeping vehicle in the collaborative job group based on its assigned subtask block and the dynamic value map, comprises: The operation control parameters comprise recommended running speed and power gear of the cleaning device, and the value of the recommended running speed and the power gear of the cleaning device are positively correlated with the cleaning value scores of corresponding road sections in the dynamic value graph.
5. The method for path planning and job management of unmanned sweeping vehicles according to claim 1, wherein the method comprises the steps of monitoring the states of all the sweeping vehicles during the execution of the job, and triggering the re-evaluation and re-planning of the tasks of the affected sweeping vehicles when a predetermined dynamic event is detected, and specifically comprises the following steps: the predetermined dynamic event comprises that the vehicle stops overtime due to faults or obstacles, the road side equipment reports a new high-priority dirt event, the remaining duration of the vehicle is insufficient or temporary traffic control information is received.
6. The method for path planning and operation management of an unmanned sweeping vehicle of claim 5, When the task re-evaluation and re-planning are triggered, if it is determined that a certain sweeper cannot continue the original task, the unfinished sub-task blocks are re-distributed to other available vehicles in the collaborative operation group or a new vehicle is scheduled to take over.
7. The method for path planning and job management of an unmanned sweeping vehicle according to claim 1, wherein after the task is completed, optimizing and iterating the calculation logic of the dynamic sweeping value score and the allocation strategy of the subtask block based on the comparison analysis result of planning data and actual execution data, specifically comprising: And calculating the plan coverage rate, the value completion rate and the path repetition rate index, and automatically adjusting the predefined strategy model parameters according to the historical data.
8. The method of path planning and job management for an unmanned motor sweeper of claim 1 wherein generating an initial global path and associated job control parameters for each motor sweeper in the collaborative job group based on its assigned subtask block and the dynamic value map further comprises: And the sweeper carries out local obstacle avoidance and track fine adjustment according to real-time sensing data of the vehicle-mounted sensor on the basis of following the initial global path.
9. The method for path planning and operation management of an unmanned sweeping vehicle of claim 1, At the beginning, a pre-stored high-precision vector map of the working area is called as a space calculation basis, wherein the high-precision vector map comprises coordinate information of lane lines, curbs and fixed facilities.
10. The path planning and operation management system of the unmanned sweeping vehicle is used for realizing the path planning and operation management method of the unmanned sweeping vehicle as described in claim 1, and is characterized by comprising a vehicle-mounted terminal, a road side sensing module and a cloud dispatching platform; The vehicle-mounted terminal is deployed on each unmanned sweeper and is used for collecting and uploading vehicle state and local environment data and receiving and executing operation instructions from the cloud; The road side sensing module is deployed at key nodes of the operation area and used for collecting and uploading wide-area traffic and environment information; The cloud scheduling platform is respectively in communication connection with the vehicle-mounted terminal and the road side perception module, wherein the cloud scheduling platform comprises: The data fusion unit is connected with the vehicle-mounted terminal and the road side perception module and is used for receiving multi-source data and generating and periodically updating a dynamic value graph; the task allocation unit is connected with the data fusion unit and is used for acquiring the dynamic value graph, decomposing a target area into a plurality of subtask blocks and allocating the subtask blocks to a group of sweeper; The path issuing unit is connected with the task distribution unit and the vehicle-mounted terminal, and is used for generating an initial global path and associated operation control parameters for each sweeper based on a distribution result and issuing the generated path and parameters to the corresponding vehicle-mounted terminal; The real-time monitoring unit is connected with the vehicle-mounted terminal, the road side sensing module and the task distribution unit and is used for monitoring the states of all the sweeper and triggering task reevaluation and rescheduling when a preset dynamic event is detected; and the data optimization unit is connected with the vehicle-mounted terminal and the data fusion unit and is used for performing optimization iteration based on the comparison analysis result of the planning data and the actual execution data.

Description

Path planning and operation management system and method for unmanned sweeper Technical Field The invention relates to the technical field of intelligent transportation, in particular to a system and a method for path planning and operation management of an unmanned sweeper. Background In the traditional urban sanitation field, cleaning operation is mainly carried out by manually driving professional vehicles or manpower, and the problems of low operation efficiency, high manpower cost, limited coverage range, difficulty in coping with night or severe weather operation and the like exist. To address these challenges, the current mainstream solution is to develop unmanned motor sweeper based on single car intelligence. The scheme is generally that a cleaning vehicle is provided with an environment sensing sensor such as a laser radar and a camera and a satellite positioning module, so that the cleaning vehicle can identify obstacles and realize basic autonomous obstacle avoidance and tracking running along a preset fixed route. Compared with complete manual operation, the technology can realize unmanned operation under certain conditions (such as a preset electronic fence or a fixed route range), so that the labor cost is primarily reduced, and the safety of basic operation is ensured. The task is executed through the preset logic, so that the problems of difficult personnel management and low standardization degree in the traditional mode are partially solved. However, the above-mentioned bicycle intelligent scheme based on a preset fixed route has a core defect in that the path planning is static and isolated. Each vehicle only independently operates according to the preset tasks, and cannot sense the global traffic state, the actual pollution degree of the operation area and the real-time position and task progress of other cooperative vehicles, so that dynamic and globally optimal path re-planning and multi-vehicle cooperative scheduling cannot be performed. The method has the advantages that the overall operation efficiency of the vehicle group is low, repeated coverage or omission is achieved, the high-efficiency cleaning requirement of sudden tasks or large-scale complex areas is difficult to deal with, and the potential of large-scale and economical deployment of the unmanned sweeper in a smart city sanitation system is severely restricted. Disclosure of Invention The invention aims to provide a path planning and operation management system and method for an unmanned sweeping vehicle, which solve the problems of static isolation and lack of cooperation of the path planning in the scheme of the existing unmanned sweeping vehicle based on single vehicle intelligence. In order to achieve the above purpose, the invention provides a path planning and operation management method of an unmanned sweeper, which comprises the following steps: acquiring multi-source data from an unmanned sweeper, road side sensing equipment and an external system, generating and periodically updating dynamic value graphs for different road sections in a working area based on the multi-source data, wherein the dynamic value graphs comprise dynamic cleaning value scores of all road sections; Decomposing a target area into a plurality of subtask blocks based on the dynamic value map, the real-time states of all available sweeper vehicles and a preset global operation target, and distributing the subtask blocks to a group of sweeper vehicles to form a cooperative operation group; Generating an initial global path and associated operation control parameters for each sweeper in the collaborative operation group based on the allocated subtask blocks and the dynamic value graphs, wherein the generation of the initial global path is prevented from generating instant spatial overlapping with the current operation areas of other sweeper in the group; During the execution process of the operation, the states of all the sweeper are monitored, and when a preset dynamic event is detected, the task reevaluation and the rescheduling of the affected sweeper are triggered; And after the task is completed, optimizing and iterating the calculation logic of the dynamic cleaning value score and the allocation strategy of the subtask block based on the comparison analysis result of the planning data and the actual execution data. And calculating the dynamic cleaning value score, wherein at least two of historical cleanliness data, a dirt event grade reported in real time, traffic flow data and weather influence factors are synthesized. The subtask block is distributed to a group of sweeper to form a cooperative operation group, and specifically comprises the following steps: And calculating based on the real-time position of the vehicle, the remaining operation endurance and the estimated time consumption of driving to each block to achieve the global operation target, wherein the global operation target is to maximize the total finished cleaning valu