Biobjective UAV/UGV Collaborative Rendezvous Planning in Persistent Intelligent Task-Based Wireless Communication

Qin Wang,Jia Wang,Junwei Tian,Hua Chen,Yu Su,Balakrishnan Nagaraj

doi:10.1155/2021/9578783

Qin Wang, Jia Wang + Show 4 more

Open Access

https://doi.org/10.1155/2021/9578783

Copy DOI

Abstract

This paper addresses a multiobjective Unmanned Aerial Vehicle (UAV) and Unmanned Ground Vehicle (UGVs) collaboration scheduling problem in which UAVs must be recharged periodically during a persistent task, such as wireless communication and making aerial panoramic VR video by UAV. The proposal is to introduce a UGV/UAV cooperative system in which UAVs should be recharged by UGVs periodically, and also, the UGVs have to visit their task points expect for the recharge points. The objective of the problem is to schedule and plan paths with the tradeoff of path length and makespan for UAVs and UGVs. The mathematical model which can be considered a combinatorial multiobjective optimization problem is presented firstly, and the solution of the problem is composed of presubsolution and postsubsolution. The multiobjective adaptive large neighborhood is hybrid with the Pareto local search (MOALP) to resolve the problem. The experimental result shows that the proposed algorithm outperforms the compared algorithms on the rendezvous planning problem for UAVs and UGVs working collaboratively in intelligent tasks.

Highlights

Unmanned Aerial Vehicles (UAVs) have been successfully used for many applications in both civil and military scenarios, such as surveillance, precision agriculture, package delivery, and facility inspection
Sunder and Rathinam [1] deployed multiple depots in the environment to refuel one UAV working in monitoring applications, the mixed integer linear programming (MILP) model is presented, and a local search heuristic was proposed for resolving the problem. [2] considered multi-UAVs and charging stations in the working scenario, and the problem is transformed into a generalized traveling salesman problem (GTSP) and resolved by a GTSP solver
We mainly focus on multiobjective path planning for UAV/UGV collaboration systems working in persistent tasks; as mentioned above, the present related works often have the following shortcomings: firstly, path planning for charging robots was focused on while seldom works considered both replenishment robots and UAVs; secondly, the charging robots were only considered mobile charging stations while these robots can complete other tasks; and thirdly, the single objective was considered in previous works while the decision-maker may need to tradeoff multiobjectives when planning paths for path planning for UAV/UGV collaboration systems

Summary

Introduction

Unmanned Aerial Vehicles (UAVs) have been successfully used for many applications in both civil and military scenarios, such as surveillance, precision agriculture, package delivery, and facility inspection. Sunder and Rathinam [1] deployed multiple depots (charging stations) in the environment to refuel one UAV working in monitoring applications, the mixed integer linear programming (MILP) model is presented, and a local search heuristic was proposed for resolving the problem. (1) The problem is formulated as mixed integer linear programming, and planning paths for UAVs and UGVs simultaneously as well as UGVs have multitask points to visit (2) The presented mathematical model is a biobjective optimization problem, which is firstly considered in this filed as far as we have known (3) A memetic algorithm which combines PLS and MOALNS is proposed for resolving the presented path planning problem (4) To resolve the problem efficiently, several heuristics and the local search procedure are developed for the algorithm.

Problem Statement

Proposed Approach

MOALNS for the Postsubsolution

Conclusions and Future Work