Swarms of Fully Actuated Agents with Model Uncertainty

Abstract

In this chapter, we consider a swarm of N agents whose dynamics evolve based on a more realistic agent dynamics model compared to the single integrator and the double integrator point mass models considered in the preceding chapters. In particular, we assume that the dynamics of the agents obey the fully actuated model

$$ M_{i}(x_{i})\ddot{x}_{i} + f_{i}(x_{i}, \dot{x}_{i}) = u_{i}, 1 \le i \le N$$

where x _i represents the position or configuration of agent i, $M_{i}(x_{i}) \in {\mathbb R}^{n \times n}$ is its mass or inertia matrix, $f_{i}(x_{i}, \dot{x}_{i}) \in {\mathbb R}^{n}$ represents the centripetal, Coriolis, gravitational effects, and additive disturbances.