Multi Objective And Multi Traveler Path Planning For Rail Mounted Logistics Robots

At present,the national cohesion is achieving the transformation of China from a manufacturing country to a manufacturing powerhouse,and China’s industrial development has always adhered to innovation driven,intelligent transformation,strengthening foundation,and green development.The country vigorously advocates the research of unmanned workshops,intelligent chemical plants,and intelligent mines.As an important component of unmanned and intelligent workshops,robots can be responsible for completing material distribution tasks within the workshop.This article focuses on the robot path planning problem in the large-scale workshop rail mounted logistics robot system,and the specific content is as follows:(1)Studied multi-objective optimization algorithms for rail mounted logistics robots.In this paper,an Reverse Elitist Fusion Genetic Algorithm(REFGA)is proposed,which integrates the idea of "real-time updating global optimal solution of particle swarm optimization and genetic operation enriching population diversity",and introduces the initial population of elite reverse learning optimization algorithm.During the iteration process,a stagnation judgment factor is set,and if the population similarity is too high,the local optimal problem is solved by increasing the mutation probability.After verification by multiple test functions,it has been proven that the REFGA algorithm has faster convergence speed and better search accuracy.(2)Studied multi-objective dynamic path planning for rail mounted logistics robots.By analyzing the operating environment of the rail mounted logistics robot,a dynamic Dijkstra algorithm that can update the solution path was designed and improved;And a conflict coordination strategy has been developed to address the issue of in orbit conflicts for multi track logistics robots,in order to avoid in orbit collision of robots.(3)Integrating UWB positioning technology with rail mounted logistics robots.The TDOA positioning algorithm using UWB positioning technology is used to establish the position equation,and the Chan algorithm is introduced to estimate the position of the position equation,achieving real-time positioning of rail mounted logistics robots in the workshop.(4)We conducted a multi-objective and multi traveling salesman path planning experiment for material delivery using a rail mounted robot.Based on the specific situation of the production workshop and combined with the idea of multiple travel agents,a workshop material distribution optimization model for multi track logistics robots was established with the objective function of minimizing the total travel path and maximizing the robot’s carrying capacity.The REFGA optimization algorithm proposed in this paper was used to solve the mathematical model.Finally,based on the specific track layout of the workshop,conduct path planning experiments for logistics robots and dynamic adjustment experiments when encountering conflicts.(5)Completed the design of a rail mounted logistics robot system.This article designs and develops a rail mounted logistics robot,which adopts a dual core architecture of "RK3588 and STM32" in its control system,which is conducive to the functional expansion of the system.The main processor RK3588 is responsible for communication,data acquisition and processing,decision-making,and other work.The coprocessor STM32 is responsible for the specific peripheral drive of the robot body,including sensors,execution mechanisms,positioning modules,and so on.