Abstract—This paper presents a novel way to bias the sampling domain of stochastic planners by learning from example plans. We learn a generative model of a planner as a function of proximity to labeled objects in the workspace. Our motivation is that certain objects in the workspace have a local influence on planning strategies, which is dependent not only on where they are but also on what they are. We introduce the concept of a Semantic Field — a region of space in which configuration sampling is modelled as a multinomial distribution described by an underlying Dirichlet distribution. We show how the field can be trained using example expert plans, pruned according to information content and inserted into a regular RRT to produce efficient plans. We go on to show that our formulation can be extended to bias the planner into producing sequences of samples which mimic the training data.

 Collision-Free Path

  • [PDF] I. Baldwin and P. Newman, “Teaching a Randomized Planner to plan with Semantic fields,” in Towards Autonomous Robotic Systems, Plymouth, UK, 2010.
    author = {Ian Baldwin and Paul Newman},
    title = {Teaching a Randomized Planner to plan with Semantic fields},
    booktitle = {Towards Autonomous Robotic Systems},
    year = {2010},
    address = {Plymouth, UK},
    month = {August},
    note = {08},
    pdf = {http://www.robots.ox.ac.uk/~mobile/Papers/BaldwinTAROS2010.pdf},
    keywords = {Learning to Plan},