Learning Sampling Distributions for Robot Motion Planning

By Brian Ichter, James Harrison, Marco Pavone. 2017

417 citations recorded by the source corpus.

A defining feature of sampling-based motion planning is the reliance on an implicit representation of the state space, which is enabled by a set of probing samples. Traditionally, these samples are drawn either probabilistically or deterministically to uniformly cover the state space. Yet, the motion of many robotic systems is often restricted to “small” regions of the state space, due to e.g. differential constraints or collision-avoidance constraints. To accelerate the planning process, it is thus desirable to devise non-uniform sampling strategies that favor sampling in those regions where an optimal solution might lie. This paper proposes a methodology for nonuniform sampling, whereby a sampling distribution is learned from demonstrations, and then used to bias sampling. The sampling distribution is computed through a conditional variational autoencoder, allowing sample generation from the latent space conditioned on the specific planning problem. This methodology is general, can be used in combination with any sampling-based planner, and can effectively exploit the underlying structure of a planning problem while maintaining the theoretical guarantees of sampling-based approaches. Specifically, on several planning problems, the proposed methodology is shown to effectively learn representations for the relevant regions of the state space, resulting in an order of magnitude improvement in terms of success rate and convergence to the optimal cost.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Control and Systems Engineering, Optimization & Training, Robotics & Control

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Control and Systems Engineering

In Optimization & Training

In Robotics & Control

Citations in the corpus

Habitat 2.0: Training Home Assistants to Rearrange their Habitat (2021, methodology): “More recent work has however begun utilizing learning and transitioning towards hybrid methods, for example learning distributions for sampling [90], using reinforcement [98] or differentiating through the optimization [99].”
Broadly-Exploring, Local-Policy Trees for Long-Horizon Task Planning (2020, background): “This also allows the embedding of policies in classical motion planning structures [23, 24, 25, 26, 27, 28, 29, 30].”
Learning a Decentralized Multi-arm Motion Planner (2020, background): “Some of these works have improved the speed of SMPs by biasing the sampling distribution of new nodes [14, 15, 16], directly predicting the next node [17], utilizing past experience with an extra lookup table module [18], or reducing the search space through discretization [5, 6, 19].”
Motion Planning Networks (2018, methodology): “A recent method [16] uses deep neural networks trained via imitation to adaptively sample the configuration space for SMP methods.”
Dynamics Learning with Cascaded Variational Inference for Multi-Step Manipulation (2019, methodology): “[10] uses conditional variational autoencoder (CVAE) [13, 28] to learn a generative model to draw collision-free samples from the action space.”
Deep Visual Reasoning: Learning to Predict Action Sequences for Task and Motion Planning from an Initial Scene Image (2020, background): “For sampling-based motion planning, [23] learn an optimal sampling distribution conditioned on the scene and the goal to speed up planning.”
Neural RRT*: Learning-Based Optimal Path Planning (2020, background): “Ichter et al. [21] propose a conditional variational autoencoder (CVAE) to learn explicit sampling distribution for robot motion planning.”