Research On Robot Optimal Path Planning Based On Discrete Particle Swarm Optimization

At present, robots have become the most representative strategic objectives inhigh-tech fields. In much key technological research on the robot, its navigationtechnology lies in the core position. Path planning for navigation is an importanttechnical issue.In recent years, as the mobile robots are widely used in military, aerospace, andother fields, and given that most of the mobile robot is battery-powered, but becauseof the limitations of battery technology, the robot's use of time is very limited.Therefore, considering the energy constraints of mobile robot in order to ensure theircontinued viability has become an important research issue. In addition, with thedevelopment of modern logistics technologies, Vehicle Routing Problem (called VRP)becomes a research hotspot issue. Meanwhile, time window constraint in the vehiclerouting problem is applied to multi-robot path planning study, which is a newexpansion and try in robot path planning.The main study in the paper proposes as follows:First of all, the particle swarm algorithm and its development process areintroduced, and by defining some basic operation rules of operations and operators,discrete particle swarm algorithm has been improved. At the same time, the algorithmis used to solve TSP, and the performance of the algorithm is analyzed, also. Thesimulation results verify that the improved algorithm is very effective.Secondly, in order to minimize energy consumption, the improved discreteparticle swarm algorithm is applied to the single robot path planning with energyconstraint, combined with the dynamic energy threshold method to achieve nearoptimal energy consumption. In this method, to consider energy constraints in thestage of mobile robot path planning is to ensure the continued survival of mobilerobot, while combining PSO search with energy constraints, thus effectively reducingthe energy consumption of the robot to complete the planning. Meanwhile, furthervalidate that the algorithm has very good searching ability and convergence efficiency.Then, Comparison experiments in the case of Single-charging Station andMulti-charging Station (for example, in the case of two charging stations) show: to increase charging stations in the right place can also help reduce energy consumption.Finally, time window constraint in the vehicle routing problem is applied tomulti-robot path planning, and then research on multi-robot optimal path planningwith the soft time windows is proposed, while coding method on multi-robot pathplanning constructs is constructed. Meanwhile, the proposed two-stage optimizationstrategy is applied to multi-robot optimal path planning with Soft Time Windows inorder to carry out solving the optimization problem. That is, in the first stage, usingordinary optimization methods solves multi-robot path and distribution of the taskpoints based on overall optimization solution; in the second stage, based on optimalsolution optimized in the first stage, to do re-optimization of the robot's pathseparately. Experiments show: this method has better feasibility and effectiveness ofsearching capability.