Data driven adaptation of physics model for use in robotics

Lead Research Organisation: University of Cambridge
Department Name: Engineering

Abstract

My project will focus on soft robotics, which is a growing area in the field of robotics. Soft body interactions are inherently hard to model and predict because traditional analytical methods are usually not applicable. I hope to develop an adaptive, data driven approach to modelling soft body interactions. One example of the real world application of such a model is in design automation - design and manufacture of components without a human designer. The reality gap - the disparity between simulated and real world behaviours - makes design automation of objects such as simple tools difficult, meaning many real world design iterations must be manufactured to optimise the solution. An accurate and adaptive model of soft body interactions could be a step towards speeding the robotic design automation

Publications

10 25 50

Studentship Projects

Project Reference Relationship Related To Start End Student Name
EP/R512461/1 01/10/2017 30/09/2021
1949869 Studentship EP/R512461/1 01/10/2017 31/03/2021 Toby Howison