Research On Path Planning And Scheduling Of Multi-Autonomous Mobile Robot Based On Cyber-Physical System Modeling

With the development of intelligent robot technology,the application of the Multi-Autonomous Mobile Robot(Multi-AMR)system is more extensive,which has the characteristics of high flexibility,real-time,and reliability.The expansion of system scale and robot types increases the complexity of the task process,putting forward higher requirements for the planning and scheduling of the multi-autonomous mobile robot.The complete scheduling process includes critical links such as map model construction,autonomous robot positioning,path planning,collision avoidance,and task assignment,and a relatively complete theoretical research system has been formed.Current research mainly focuses on scheduling sub-processes,including algorithm design and implementation for specific constraint indexes.Considering the algorithm complexity and task requirements,system modeling,planning,and complete scheduling of the multi-autonomous mobile robot have become new research hotspots.In recent years,the modeling research of complex robot systems based on the Cyber-Physical System(CPS)has become more mature.The CPS effectively realizes the information coupling and independent deployment of physical actuators,communication processes,and control algorithms as a fully autonomous or semi-autonomous model.In addition,technologies such as big data and edge computing have emerged,and CPS-based scheduling of the multi-autonomous mobile robot has become a new research direction,achieving optimization of constraint indexes,computational efficiency,and task efficiency.This topic studies the path planning and scheduling strategy of the multi-autonomous mobile robot based on the CPS.A map construction and positioning method based on a two-dimensional topological grid is proposed,the heuristic path planning algorithm with time and space constraints is studied,collision avoidance and task assignment algorithms for sharing map information are constructed,and the complete scheduling process and algorithm deployment of the multi-autonomous mobile robot is realized.The research contents and main contributions are as follows:(1)Two-dimensional map modeling and vehicle-mounted visual positioning methods for multi-autonomous mobile robot systems are studied.The path state evaluation index of path arc time cost is introduced,and a proposed two-dimensional topological grid map construction method.The optimal expected value is solved using probability theory,and the evaluation of the path congestion state is realized.An improved two-dimensional feature coding positioning landmark and vehicle-mounted visual sensor positioning method are proposed,which realizes the continuous positioning and position prediction of autonomous mobile robots during the movement,and avoids the influence of motion blur on visual positioning.Further,the reliability of the improved map and visual positioning method is analyzed based on the CPS architecture.(2)A time-space constrained A* path planning method is studied for the multi-autonomous mobile robot under the CPS.According to the topological grid map model,a Time-Constrained A*(TCA*)algorithm based on path arc time cost is proposed,and a hierarchical planning process including path evaluation and generation is designed to realize the congestion evaluation and avoidance.A Spatio-Temporal Constrained A*(STCA*)algorithm based on path arc time cost and destination node distance is proposed,and time and space constraints are introduced to enhance the motion order of autonomous mobile robots.Combined with the CPS model,distributed path planning and deployment are realized in the physical layer,improving computing and task efficiency.(3)Node information-based collision avoidance and space-time-priority-constrained task assignment methods for the multi-autonomous mobile robot are studied.Based on the node occupancy status of the topological grid map,a priority list of all nodes is established,an orderly collision avoidance process is realized,and the system congestion and deadlock are alleviated.A task assignment strategy with space-time-priority constraints is proposed by synthesizing the path arc time cost,destination node distance,and task priority.By pre-evaluating the task status,the order and timeliness of the task access process are optimized,and the congestion around the entrance with a high task load is alleviated.Under the CPS,the collaborative reliability of collision avoidance,task assignment,and path planning is verified,respectively.(4)The multi-autonomous mobile robot comprehensive scheduling algorithm and performance evaluation are studied.The path cost normal statistics,global task decision-making,and collision avoidance methods are achieved in the control layer.Combined with autonomous visual positioning,distributed path planning,and task evaluation strategies in the physical layer,a system architecture is realized with efficient information exchange and balanced computing power.Key evaluation indicators are proposed,including task efficiency,average parking time,average planning time,and deadlock frequency,used to evaluate system performance.The general architecture of the multi-autonomous mobile robot scheduling based on CPS modeling improves overall task efficiency and system security.(5)The application of the multi-autonomous mobile robot in a self-service check-in system under the CPS modeling is studied.The check-in island topological grid map is established.The complete scheduling of the multi-check-in autonomous mobile robot is deployed and realized under the improved CPS modeling framework,including the two-dimensional map status update and global collision avoidance in the control layer and the space-time constrained path planning and task assignment in the physical layer.Experiments show that the improved scheme dramatically improves task efficiency in practical applications.In summary,this research focuses on CPS modeling and multi-autonomous mobile robot scheduling,analyzes and verifies the implementation method of the scheduling sub-process,and proposes a general architecture for multi-autonomous mobile robot scheduling under CPS modeling.The results have theoretical and practical significance for improving the task efficiency,system security,and cyber-physical system model management of complex robot collaboration.