Fast and Compute-efficient Sampling-based Local Exploration Planning via Distribution Learning
arXiv:2202.13715 · doi:10.1109/LRA.2022.3186511
Abstract
Exploration is a fundamental problem in robotics. While sampling-based planners have shown high performance, they are oftentimes compute intensive and can exhibit high variance. To this end, we propose to directly learn the underlying distribution of informative views based on the spatial context in the robot's map. We further explore a variety of methods to also learn the information gain. We show in thorough experimental evaluation that our proposed system improves exploration performance by up to 28% over classical methods, and find that learning the gains in addition to the sampling distribution can provide favorable performance vs. compute trade-offs for compute-constrained systems. We demonstrate in simulation and on a low-cost mobile robot that our system generalizes well to varying environments.
Accepted for IEEE RA-L. Open-source code: https://github.com/ethz-asl/cvae_exploration_planning, 8 pages, 12 figures
References in corpus (6)
- Voxblox: Incremental 3D Euclidean Signed Distance Fields for On-Board MAV Planning
- An Efficient Sampling-based Method for Online Informative Path Planning in Unknown Environments
- Fast Frontier-based Information-driven Autonomous Exploration with an MAV
- History-aware Autonomous Exploration in Confined Environments using MAVs
- Panoptic Multi-TSDFs: a Flexible Representation for Online Multi-resolution Volumetric Mapping and Long-term Dynamic Scene Consistency
- A Unified Approach for Autonomous Volumetric Exploration of Large Scale Environments under Severe Odometry Drift