Research On Path Planning And Visualization Of Multi-Mobile Robot System With Energy Consumption Constraint

With the rapid development of the artificial intelligence industry,mobile robots are now widely used in manufacturing,service industry,aerospace,medical and other civilian and military fields,etc.They can replace humans working in harsh and dangerous environments,etc.Multiple mobile robots share data and work together to greatly improve the efficiency of completing tasks that cannot be done by a single mobile robot.There are some typical application requirements,for example,smart factories delivering goods from warehouses to different processing stations and courier companies delivering deliveries from business locations to different buildings in residential communities.Mobile robots are usually powered by batteries,which have the limitation of providing limited energy.This study focuses on energy-optimized multi-mobile robot path planning,multi-mobile robot power replacement scheduling,and the virtual reality visualization.(1)The multi-mobile robot collision avoidance problem is studied.This study proposes a dynamic multi-mobile robot collision avoidance(TW&BT)fusion algorithm that is based on rolling time windows and binary tree prior traversal under energy optimization.This algorithm is based on the improved A-star algorithm to find the optimal initial path under the energy consumption constraint.Based on the cooperative mechanism of rolling time window and binary tree prior traversal,the whole operation time axis is decomposed into multiple time windows which takes the mobile robot collision as the trigger event in the initial path.During each time window,the optimal collision avoidance decision is solved by the algorithm which is based on binary tree prior traversal with the objective of minimizing energy consumption when stopping to avoid collision.The simulation results show that the TW&BT fusion algorithm has high robustness.To compare with the dynamic priority-based conflict resolution strategy(DPS)method,the TW&BT fusion algorithm achieves collision avoidance while reducing the generated energy consumption by 33.1% within a similar computation time.(2)The optimization problem of selecting and scheduling battery swapping stations based on energy consumption optimization is studied.This study proposes an improved genetic algorithm(GA~*)for selecting and scheduling battery swapping stations which is based on the need to replace power sources during mobile robot operations.A multi-objective model is established to consider the energy consumption cost of the actual driving of the mobile robot,the cost of setting up the battery swapping station,the cost of waiting for the replacement of the power supply and other multi-objective constraints.The multi-objective model of the battery swapping station location scheduling based on energy consumption optimization is established.For introducing the iterative mechanism of the second cycle of the genetic population,the results are solved by the proposed GA~* for the battery swapping station location scheduling optimization with energy consumption optimization.To compare with the traditional GA,the simulation experiments show that the GA~* algorithm has high effectiveness,robustness and timeliness.(3)The visualization problem of the path planning of a multimobile robot system is studied.Based on Unity3 D,it can realize the virtual reality visualization of the path planning results of a multi-mobile robot system under the energy consumption constraint.The simulated the experimental map is built.It includes a virtual environment with obstacles for robot operation,and a battery swapping station set up based on the GA~*algorithm.The visualization platform not only dynamically visualizes the mobile robots moving and exchanging electricity according to the path found by the TW&BT fusion algorithm and draws the movement route,but also displays the energy consumption of the mobile robots in real time.Finally,the immersive roaming of the mobile robot in a virtual operating environment with obstacle avoidance motion is realized in the Virtual Reality Engineering Centre.