Perceptual Attention-based Predictive Control

TL;DR

This paper introduces a perceptual attention-based predictive control system that uses model predictive control and neural networks to improve safety in autonomous visual navigation by early detection of unsafe conditions.

Contribution

It presents a novel architecture combining MPC, CNNs, and uncertainty quantification for attention-based visual control in autonomous systems, enhancing safety detection capabilities.

Findings

Outperforms previous methods in early unsafe condition detection

Uses uncertainty estimates to assess safety in real-time

Validated on a scaled terrestrial vehicle

Abstract

In this paper, we present a novel information processing architecture for safe deep learning-based visual navigation of autonomous systems. The proposed information processing architecture is used to support a perceptual attention-based predictive control algorithm that leverages model predictive control (MPC), convolutional neural networks (CNNs), and uncertainty quantification methods. The novelty of our approach lies in using MPC to learn how to place attention on relevant areas of the visual input, which ultimately allows the system to more rapidly detect unsafe conditions. We accomplish this by using MPC to learn to select regions of interest in the input image, which are used to output control actions as well as estimates of epistemic and aleatoric uncertainty in the attention-aware visual input. We use these uncertainty estimates to quantify the safety of our network controller under the current navigation condition. The proposed architecture and algorithm is tested on a 1:5 scale terrestrial vehicle. Experimental results show that the proposed algorithm outperforms previous approaches on early detection of unsafe conditions, such as when novel obstacles are present in the navigation environment. The proposed architecture is the first step towards using deep learning-based perceptual control policies in safety-critical domains.