Nonlinear MPC for Collision Avoidance and Controlof UAVs With Dynamic Obstacles (2008.00792v1)

Abstract: This article proposes a Novel Nonlinear Model Predictive Control (NMPC) for navigation and obstacle avoidance of an Unmanned Aerial Vehicle (UAV). The proposed NMPC formulation allows for a fully parametric obstacle trajectory, while in this article we apply a classification scheme to differentiate between different kinds of trajectories to predict future obstacle positions. The trajectory calculation is done from an initial condition, and fed to the NMPC as an additional input. The solver used is the nonlinear, non-convex solver Proximal Averaged Newton for Optimal Control (PANOC) and its associated software OpEn (Optimization Engine), in which we apply a penalty method to properly consider the obstacles and other constraints during navigation. The proposed NMPC scheme allows for real-time solutions using a sampling time of 50 ms and a two second prediction of both the obstacle trajectory and the NMPC problem, which implies that the scheme can be considered as a local path-planner. This paper will present the NMPC cost function and constraint formulation, as well as the methodology of dealing with the dynamic obstacles. We include multiple laboratory experiments to demonstrate the efficacy of the proposed control architecture, and to show that the proposed method delivers fast and computationally stable solutions to the dynamic obstacle avoidance scenarios.

Nonlinear Model Predictive Control for UAV Collision Avoidance

The paper "Nonlinear MPC for Collision Avoidance and Control of UAVs with Dynamic Obstacles" presents a Nonlinear Model Predictive Control (NMPC) framework specifically tailored for the navigation and dynamic obstacle avoidance of Unmanned Aerial Vehicles (UAVs). The paper explores the application of NMPC, a control strategy that predicts and optimizes future system behavior over a given horizon, in UAV contexts where dynamic obstacles are prevalent.

NMPC Framework and Methodology

The NMPC framework utilizes the Proximal Averaged Newton-type method for Optimal Control (PANOC) as the solver, which incorporates the Optimization Engine (OpEn) software. This solver is selected due to its efficiency in handling nonlinear, non-convex optimization problems and its capability to generate solutions with low computational footprint in real-time scenarios. The approach accounts for the nonlinear dynamics of UAVs by leveraging kinematic models that consider gravitational and aerodynamic forces, as well as compensation for roll and pitch inputs. The control problem involves minimizing a cost function designed to ensure state and input reference tracking while enforcing smooth and feasible control actions.

Dynamic Obstacle Handling

A significant contribution of this paper is the integration of dynamic obstacle trajectories within the NMPC framework. Obstacles are not presumed static, and the trajectory prediction is incorporated into the control loop. The prediction model admits different types of trajectories including linear and projectile motion. The trajectory classification scheme compares backwards-integrated predictions to recent measurements to determine the best-fit model for obstacle motion, enhancing the prediction accuracy.

Experimental Validation

The efficacy of the proposed control architecture is validated through a series of laboratory experiments using controlled environments and motion capture systems. When subjected to dynamic environments with moving obstacles, the NMPC framework demonstrated robust avoidance capabilities. The UAV successfully avoided collision paths with thrown projectiles and moving pedestrian-like obstacles, showcasing the real-time adaptability and computational stability of the approach.

Implications and Future Research

This research illustrates the potential for NMPC in complex UAV navigation tasks involving dynamic obstacles. From a practical standpoint, such frameworks are pivotal for deploying UAVs in urban environments or other settings where interaction with dynamic entities cannot be pre-determined or statically modeled. Theoretically, insights from this paper may influence future developments in NMPC algorithms, especially in enhancing solver efficiency and handling more complex, unpredictable trajectories.

Although the trajectory prediction method marks substantial progress, the reliance on a motion capture system limits real-world applicability. Future research could explore onboard sensing and prediction using stereo-cameras or lidars, thereby increasing autonomy and versatility. Additionally, expanding the trajectory classification scheme to include more diverse dynamics or integrating with machine learning techniques for adaptive learning could further bolster the robustness and applicability of NMPC-based UAV control systems in dynamic, real-world environments.