🤖 OMEGA: Efficient Occlusion-Aware Navigation for Air-Ground Robot in Dynamic Environments via State Space Model

1The University of Hong Kong (HKU)
2Shanghai AI Laboratory
3Huazhong University of Science and Technology
Submitted to IEEE RA-L 2024 (Under Review)

*denotes corresponding author.



To the best of our knowledge, OMEGA is the first system tailored for AGR to efficiently navigate in dynamic environments. It combines OccMamba and AGR-Planner to ensure fast and efficient autonomous navigation in highly occluded and rapidly changing environments.

Abstract

Air-ground robots (AGRs) are widely used in surveillance and disaster response due to their exceptional mobility and versatility (i.e., flying and driving). Current AGR navigation systems perform well in static occlusion-prone environments (e.g., indoors) by using 3D semantic occupancy networks to predict occlusions for complete local mapping and then computing Euclidean Signed Distance Field (ESDF) for path planning. However, these systems face challenges in dynamic, severe occlusion scenes (e.g., crowds) due to limitations in perception networks' low prediction accuracy and path planners' high computation overhead.

In this paper, we propose OMEGA, which contains OccMamba with an Efficient AGR-Planner to address the above-mentioned problems. OccMamba adopts a novel architecture that separates semantic and occupancy prediction into independent branches, incorporating two mamba blocks within these branches. These blocks efficiently extract semantic and geometric features in 3D environments with linear complexity, ensuring that the network can learn long-distance dependencies to improve prediction accuracy. Semantic and geometric features are combined within the Bird's Eye View (BEV) space to minimise computational overhead during feature fusion. The resulting semantic occupancy map is then seamlessly integrated into the local map, providing occlusion awareness of the dynamic environment. Our AGR-Planner utilizes this local map and employs kinodynamic A* search and gradient-based trajectory optimization to guarantee planning is ESDF-free and energy-efficient.

Extensive experiments demonstrate that OccMamba outperforms the state-of-the-art 3D semantic occupancy network with 25.0% mIoU. End-to-end navigation experiments in dynamic scenes verify OMEGA's efficiency, achieving a 96% average planning success rate. Code and video are available at https://jmwang0117.github.io/OMEGA/.

Experiments

OMEGA = OccMamba + AGR-Planner

Key Component: OccMamba

OMEGA vs. Baselines

Real Dynamic Scene Navigation

OMEGA Real World Experiment

OMEGA Real World Experiment

Simulated Dynamic Scene Navigation

OMEGA Simulation Experiment

OMEGA Simulation Experiment

Video Presentation

BibTeX


        @article{wang2024omega,
          title={OMEGA: Efficient Occlusion-Aware Navigation for Air-Ground Robot in Dynamic Environments via State Space Model},
          author={Wang, Junming and Huang, Dong and Guan, Xiuxian and Sun, Zekai and Shen, Tianxiang and Liu, Fangming and Cui, Heming},
          journal={arXiv preprint arXiv:2408.10618},
          year={2024}
        }