DeepAI AI Chat
Log In Sign Up

Learning Obstacle Representations for Neural Motion Planning

08/25/2020
by   Robin Strudel, et al.
0

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.

READ FULL TEXT

page 2

page 7

12/12/2022

Informed Circular Fields for Global Reactive Obstacle Avoidance of Robotic Manipulators

In this paper a global reactive motion planning framework for robotic ma...
05/03/2022

Sampling-free obstacle gradients and reactive planning in Neural Radiance Fields (NeRF)

This work investigates the use of Neural implicit representations, speci...
07/15/2019

Sampling-based Motion Planning via Control Barrier Functions

Robot motion planning is central to real-world autonomous applications, ...
08/26/2021

Parallelised Diffeomorphic Sampling-based Motion Planning

We propose Parallelised Diffeomorphic Sampling-based Motion Planning (PD...
06/15/2023

Guided Sampling-Based Motion Planning with Dynamics in Unknown Environments

Despite recent progress improving the efficiency and quality of motion p...
09/02/2023

A Unifying Variational Framework for Gaussian Process Motion Planning

To control how a robot moves, motion planning algorithms must compute pa...
09/14/2022

Uncertainty-Aware Visual Perception for Safe Motion Planning

For safe operation, a robot must be able to avoid collisions in uncertai...