Coevolution Pigeon-Inspired Optimization with Cooperation-Competition Mechanism for Multi-UAV Cooperative Region Search

Abstract

In this paper, a dynamic two-stage closed search (DTSCS) scheme for the unmanned aerial vehicle (UAV) cooperative region search is designed, which satisfies the range constraint (RC) and orientation constraint (OC). The closed trajectory is composed of two coupling stages, the search stage and the return stage. The position and orientation at the end of the search stage are the starting cell and orientation of the return stage. In the first stage, a coevolution pigeon-inspired optimization (CPIO) algorithm based on the cooperation-competition mechanism is proposed for multi-UAV cooperative search. In the return stage, inspired by region searching and trajectory tracking, a search tracking (ST) approach is presented to obtain the lowest-cost path under OC. The simulation results show that: (i) N p = 5 is the best prediction time step. (ii) CPIO algorithm performs better than the compared intelligent algorithms in region searching. (iii) ST has high tracking performance than other algorithms. (iv) The DTSCS scheme enables every UAV to make the best use of its fuel to cover more region and return to the airport within the RC, and the average range utilization of UAVs is 97% under the 3OC.

A review of cooperative path planning of an unmanned aerial vehicle group

“…The circle packing search algorithm was used to minimize the number of repeated searches of the explored region, while the UAV trajectories were generated using Dubins path planning. Luo et al (2019) studied the cooperative search for a UAV group in an uncertain 2D environment, and proposed a co-evolution pigeoninspired optimization (CPIO) algorithm based on the cooperation-competition mechanism to maximize the target existence probability and to minimize the environmental uncertainty. Then, the search tracking approach was used to ensure that the UAVs safely returned to the starting base.…”

Cooperative Path Planning for Multiple Robots With Motion Constraints in Obstacle-Strewn Environment

“…About the region searching knowledge [2]: The cell with a high probability of target existence is modeled as the key region. The cell with a low probability of target existence is modeled as the non-key region, and the no-fly zone is regarded as the obstacle.…”

“…We extend the application object of the search tracking approach proposed in [2] to multiple robots and design the CST approach to track the grid paths under collision-free and motion constraints. Furthermore, according to the scheme of solving CPPMR, a CPS approach coupled by the CP method and the CST approach is desinged.…”

“…In a busy logistics warehouse, it is not only required to minimize the total travel distance of all robots from their specified starting points to the goal points but also to avoid collisions between robots. Besides, every robot should abide by motion constraints [2] and have the ability to cope with unexpected events in obstacle-strewn environment. CPPMR has been proved to be a NP-hard problem in [1], [3], and seeking efficient schemes is always the research hot topic in recent decades.…”

“…(iii) Besides the primary constraint that robots cannot collide with each other, there are also some important conditions in the actual scene. Such as motion constraints [2] and temporal logic constraint [27].…”