Sparse Identification of Nonlinear Dynamics-based Model Predictive Control for Multirotor Collision Avoidance

TL;DR

The paper tackles collision avoidance for multirotors when payload-induced dynamics create model uncertainty. It introduces a data-driven framework, SINDy-MPC, that learns a nominal nonlinear model \hat{f} from limited data and integrates it into an MPC with obstacle-avoidance constraints over horizon $N$. Through MATLAB simulations, translational and rotational dynamics are identified with small parameter errors, and SINDy-MPC outperforms nominal MPC under $20\%$ mass uncertainty and unknown aerodynamics, achieving accurate trajectory tracking and obstacle avoidance. The approach offers a practical, real-time capable solution for robust autonomous flight with payload variability, with hardware validation left for future work.

Abstract

This paper proposes a data-driven model predictive control for multirotor collision avoidance considering uncertainty and an unknown model from a payload. To address this challenge, sparse identification of nonlinear dynamics (SINDy) is used to obtain the governing equation of the multirotor system. The SINDy can discover the equations of target systems with low data, assuming that few functions have the dominant characteristic of the system. Model predictive control (MPC) is utilized to obtain accurate trajectory tracking performance by considering state and control input constraints. To avoid a collision during operation, MPC optimization problem is again formulated using inequality constraints about an obstacle. In simulation, SINDy can discover a governing equation of multirotor system including mass parameter uncertainty and aerodynamic effects. In addition, the simulation results show that the proposed method has the capability to avoid an obstacle and track the desired trajectory accurately.

Figures (10)

• Figure 1: Configuration of multirotor with payload.
• Figure 2: Concept of SINDy.
• Figure 3: Schematic of SINDy-MPC.
• Figure 4: Schematic of control mode to collect data.
• Figure 5: Concept of model predictive control.