Planning answers the question “How should I act?” This is an essential part of the autonomy pipeline as these algorithms govern how we interact with the world.

This is quite difficult as the planning problem makes use of the outputs from localisation, mapping and perception, and imposes additional challenges such as dynamic obstacles, vehicle limitations and real-time constraints. These algorithms must be reliable for platforms spanning from a few kilograms to many-tonne vehicles.

In addition to conventional motion planning, we are investigating other planning applications such as exploration for mapping, reducing energy consumption and learning from expert demonstration.


Our latest research

Teaching a Randomized Planner to plan with Semantic fields

 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 ...
Read More

Planning Most-Likely Paths from Overhead Imagery

In this work, we are concerned with planning paths from overhead imagery. The novelty here lies in taking explicit account of uncertainty in terrain classification and spatial variation in terrain cost. The image is first classified using a multi-class Gaussian Process Classifier which ...
Read More

Non-parametric Learning for Natural Plan Generation

We present a novel way to learn sampling distributions for sampling-based motion planners by making use of expert data. We learn an estimate (in a non-parametric setting) of sample densities around semantic regions of interest, and incorporate these learned distributions ...
Read More

Planning to Perceive: Exploiting Mobility For Robust Object Detection

Abstract - Consider the task of a mobile robot autonomously navigating through an environment while detecting and mapping objects of interest using a noisy object detector. The robot must reach its destination in a timely manner, but is rewarded for correctly ...
Read More

Risky Planning: Path Planning over Costmaps with a Probabilistically Bounded Speed-Accuracy Tradeoff

Abstract— This paper is about generating plans over uncertain maps quickly. Our approach combines the ALT (A* search, landmarks and the triangle inequality) algorithm and risk heuristics to guide search over probabilistic cost maps. We build on previous work which ...
Read More

Advanced BIT* (ABIT*): Sampling-Based Planning with Advanced Graph-Search Techniques

Advanced BIT* (ABIT*): Sampling-Based Planning with Advanced Graph-Search Techniques Path planning is the problem of finding a continuous sequence of valid states from a start to a goal specification. Popular approaches in robotics include graph-based searches, such as A* [1], ...
Read More