Agile Formation Control for Intelligent Swarm Systems With Guaranteed Collision Avoidance

• Published in: IEEE transactions on intelligent transportation systems 2024-11, Vol.25 (11), p.18501-18514
• Main Authors: Wei, Cui, Chen, Ye-Hwa, Chai, Tianyou, Fu, Jun
• Format: Article
• Language: English
• Subjects: Agile formation; Collaboration; Collision avoidance; constraint following; Formation control; Kinematics; Particle swarm optimization; swarm systems; Switches; Vehicle dynamics
• ISSN: 1524-9050; 1558-0016
• DOI: 10.1109/TITS.2024.3405349

Agile formation (AF) is a new frontier for intelligent swarm system formation. The AF pertains to perform various tasks in short phases of work and frequent reassessment and adaptation of plans. This greatly increases the applicability of swarm systems. There are however two major challenges for the control design: smooth task transitions and guaranteed collision avoidance. We adopt the constraint-following approach to address these. First, for agile formation, a plateau activation function is proposed to generate a sequence of consecutive and disjoint formations. For collision avoidance, a distance-gauge function is proposed. Second, by taking the objectives of agile formation control and collision avoidance as desirable constraints, the agile formation together with collision avoidance are both cast into a constraint following control problem. Third, to evaluate the constraint-following error, a performance measure \beta is introduced and then an agile formation control is designed to render the \beta -measure to be asymptotically convergent to zero. By this, the swarm system can follow the agile formation constraint and collision avoidance constraint. Therefore, agile formation and collision avoidance are both accomplished.