Robot

Algorithm

Materials science

Sorting

Obstacle avoidance

Computer science

Mathematical optimization

Distributed computing

Mathematics

Genetic algorithm

Mobile robot

Artificial intelligence

Flocking (texture)

Composite material

Optimization problem

Machine learning

Information Sciences

Unmanned aerial vehicle (UAV) flocking control with obstacle avoidance is a many-objective optimization problem for centralized algorithms. A UAV flocking distributed optimization control frame is designed to render the many-objective optimization problem into a multi-objective optimization solved by a single UAV. For different objectives, two kinds of criteria are raised to guarantee flight safety: the hard constraints that must be satisfied and the soft ones that will be optimized. Considering the restrictions of onboard computing resources, multi-objective pigeon-inspired optimization (MPIO) is modified based on the hierarchical learning behavior in pigeon flocks. On such a basis, a UAV distributed flocking control algorithm based on the modified MPIO is proposed to coordinate UAVs to fly in a stable formation under complex environments. Comparison experiments with basic MPIO and a modified non-dominated sorting genetic algorithm (NSGA-II) are carried out to show the feasibility, validity, and superiority of the proposed algorithm.

A multi-objective pigeon-inspired optimization approach to UAV distributed flocking among obstacles