Learning Obstacle Representations for Neural Motion Planning
This work addresses the problem of robust motion planning for robotics in unfamiliar settings, representing an incremental advance through a novel hybrid method.
The paper tackles sensor-based motion planning in new and dynamic environments by learning obstacle representations using a PointNet-based architecture, achieving significant improvements in accuracy and efficiency over state-of-the-art methods.
Motion planning and obstacle avoidance is a key challenge in robotics applications. While previous work succeeds to provide excellent solutions for known environments, sensor-based motion planning in new and dynamic environments remains difficult. In this work we address sensor-based motion planning from a learning perspective. Motivated by recent advances in visual recognition, we argue the importance of learning appropriate representations for motion planning. We propose a new obstacle representation based on the PointNet architecture and train it jointly with policies for obstacle avoidance. We experimentally evaluate our approach for rigid body motion planning in challenging environments and demonstrate significant improvements of the state of the art in terms of accuracy and efficiency.