Neural NMPC through Signed Distance Field Encoding for Collision Avoidance

TL;DR

This paper presents a neural NMPC framework that uses a neural network to encode environment information into an SDF for collision-free navigation of aerial robots in unknown, cluttered environments, with theoretical guarantees and real-world validation.

Contribution

It introduces a novel neural architecture for environment encoding into SDFs integrated with NMPC, providing stability analysis and real-world validation for collision avoidance.

Findings

Effective collision avoidance in cluttered environments

Robustness to drifting odometry and adversarial inputs

Successful real-world forest navigation experiments

Abstract

This paper introduces a neural Nonlinear Model Predictive Control (NMPC) framework for mapless, collision-free navigation in unknown environments with Aerial Robots, using onboard range sensing. We leverage deep neural networks to encode a single range image, capturing all the available information about the environment, into a Signed Distance Function (SDF). The proposed neural architecture consists of two cascaded networks: a convolutional encoder that compresses the input image into a low-dimensional latent vector, and a Multi-Layer Perceptron that approximates the corresponding spatial SDF. This latter network parametrizes an explicit position constraint used for collision avoidance, which is embedded in a velocity-tracking NMPC that outputs thrust and attitude commands to the robot. First, a theoretical analysis of the contributed NMPC is conducted, verifying recursive feasibility and stability properties under fixed observations. Subsequently, we evaluate the open-loop performance of the learning-based components as well as the closed-loop performance of the controller in simulations and experiments. The simulation study includes an ablation study, comparisons with two state-of-the-art local navigation methods, and an assessment of the resilience to drifting odometry. The real-world experiments are conducted in forest environments, demonstrating that the neural NMPC effectively performs collision avoidance in cluttered settings against an adversarial reference velocity input and drifting position estimates.