Control of a Robotic Swarm Formation to Track a Dynamic Target with Communication Constraints: Analysis and Simulation

Abstract

We describe and analyze the Local Charged Particle Swarm Optimization (LCPSO) algorithm, that we designed to solve the problem of tracking a moving target releasing scalar information in a constrained environment using a swarm of agents. This method is inspired by flocking algorithms and the Particle Swarm Optimization (PSO) algorithm for function optimization. Four parameters drive LCPSO—the number of agents; the inertia weight; the attraction/repulsion weight; and the inter-agent distance. Using APF (Artificial Potential Field), we provide a mathematical analysis of the LCPSO algorithm under some simplifying assumptions. First, the swarm will aggregate and attain a stable formation, whatever the initial conditions. Second, the swarm moves thanks to an attractor in the swarm, which serves as a guide for the other agents to head for the target. By focusing on a simple application of target tracking with communication constraints, we then remove those assumptions one by one. We show the algorithm is resilient to constraints on the communication range and the behavior of the target. Results on simulation confirm our theoretical analysis. This provides useful guidelines to understand and control the LCPSO algorithm as a function of swarm characteristics as well as the nature of the target.

Balancing Collective Exploration and Exploitation in Multi-Agent and Multi-Robot Systems: A Review

“…Stogiannos et al (2020) measured the level of exploration and exploitation by tracking three metrics: 1) the minimum distance between agent pairs, 2) the sum of the distances between detected targets and their closest agents, and 3) the distance moved by each agent over the course of one time-step. While often not explicitly stated, the distance between a robot and its neighbors is the most common measure of diversity as most control methods tend to focus on preventing excessive aggregation or spatial distribution (Hereford and Siebold, 2010;Meyer-Nieberg et al, 2013;Lv et al, 2016;Meyer-Nieberg, 2017;Chen and Huang, 2019;Coquet et al, 2019;Coquet et al, 2021).…”

“…These clusters were also distributed across the search space, demonstrating a balance between exploration and exploitation, allowing the system to capture fast-moving targets. By combining and adjusting the strength of separate inter-agent attraction and repulsion fields Coquet et al (2021), has demonstrated that a stable equilibrium position can be attained where agents maintain a fixed relative position to each other even though the entire swarm may be in motion. This allowed for the overall surface of an MRS, and hence the Okumura et al (2018) Varying time interval at which robots regroup to trade map information Area Exploration Schumer and Steiglitz (1968), Azad and Hasançebi (2014), Hansen and Ostermeier (2001), Hansen (2006) Adaptive step size Optimization Blackwell and Bentley (2002) Exponential inter-agent repulsion strength Optimization Kernbach et al (2009), Schmickl et al (2009), Bodi et al (2012), Hereford (2013), Bodi et al (2015), Kengyel et al (2016) 2019), Coquet et al (2021), Kwa et al (2020a), Kwa et al (2020b), Kwa et al (2021a), Kwa et al (2021b) Exponential inter-agent repulsion and attraction strength Target Tracking Parker and Emmons (1997), Parker (2002), Kolling and Carpin (2006), Kolling and Carpin (2007) Linear inter-agent repulsion strength & variable target attraction strength Target Tracking (CMOMMT) overall EED of the swarm, to be controlled even though the entire system may be moving.…”

Model Predictive Control for Cooperative Transportation with Feasibility-Aware Policy

“…Because of these requirements, MPC approaches remain preferable [40]. Moreover, while in most of the MPC literature, collision avoidance is included in the problem formulation as hard constraints, some authors adopt the Potential Field Method [41] for trajectory generation, which can be easily adapted for both the leader and the follower and can deal also with additional communication constraints [42]. Local minima may appear with this latter approach, where the robots get stuck easily in presence of many obstacles, but recent advances propose ways to overcome this drawback [43].…”

Collective Decision Making in Communication-Constrained Environments

“…Limited range and spatially targeted communications have also been proposed that are design decisions to improve the collective behaviour rather than an externally-imposed constraint [10], [11], [12]. The resilience of different control approaches to communication noise and temporal constraints are considered [13], [14] but, to the best of our knowledge, literature lacks a collective decision-making strategy that adapts to environments with inhomogeneous communication quality distribution.…”