Learning-based 3D Occupancy Prediction for Autonomous Navigation in Occluded Environments

TL;DR

This paper introduces a deep learning approach for predicting occupancy in occluded spaces to improve autonomous navigation safety and efficiency in cluttered environments.

Contribution

It presents a novel neural network model that predicts obstacle distributions from partial observations, enabling real-time, safe navigation without ground-truth data.

Findings

Improves safety without reducing navigation speed.

Effective in unseen, cluttered environments.

Utilizes unlabeled data for training.

Abstract

In autonomous navigation of mobile robots, sensors suffer from massive occlusion in cluttered environments, leaving significant amount of space unknown during planning. In practice, treating the unknown space in optimistic or pessimistic ways both set limitations on planning performance, thus aggressiveness and safety cannot be satisfied at the same time. However, humans can infer the exact shape of the obstacles from only partial observation and generate non-conservative trajectories that avoid possible collisions in occluded space. Mimicking human behavior, in this paper, we propose a method based on deep neural network to predict occupancy distribution of unknown space reliably. Specifically, the proposed method utilizes contextual information of environments and learns from prior knowledge to predict obstacle distributions in occluded space. We use unlabeled and no-ground-truth data to train our network and successfully apply it to real-time navigation in unseen environments without any refinement. Results show that our method leverages the performance of a kinodynamic planner by improving security with no reduction of speed in clustered environments.