Skip to main content

Bio-Robotic Control

Bio-Robotic Control

Academic Contact: Netta Cohen
Academic Staff: Dr Graham Askew, Dr Samit Chakrabarty, Professor Abbas A. Dehghani-Sanij, Professor Mark Mon-Williams, Professor Netta Cohen, Professor Robert Richardson, Professor Shane Xie

How do we control our behaviour, and how should we design control systems for autonomous robots? These two questions are often approached using vastly different approaches and methodologies. Research and Development in Biorobotic Control sits around the interface between biological (animal, plant and even cellular) and robotic control, using the language of control to better understand natural systems, and using our understanding of biological systems to inspire, inform and design the robotic control systems of tomorrow.

Activity in this area is closely linked with basic research in:

  • Computational and Systems Neuroscience
  • Biomechanics, Biological Physics and Physiology

It is applied across a range of domains including:

  • Exploration Robotics
  • Infrastructure Robotics
  • Surgical Robotics

Adaptive Control of Undulatory Locomotion

An illustrative example of research on Bio-Robotic Control at Leeds is a body of work on undulatory (snake-like) locomotion. As is often the case in this field, the work began with a detailed study of a particular organism – the nematode worm Caenorhabditis elegans. This small and unassuming creature is one of the most widely studied model organisms, and crawls using side-to-side undulations much like a snake. Jordan Boyle, Stefano Berri, Ian Hope and Netta Cohen developed a novel experimental methodology that shed new light on how the worm’s locomotion is modulated by the properties of the environment:

This informed the development of an integrated neuro-mechanical model by Boyle and Cohen, which was ultimately able to replicate the generation and modulation of the worm’s locomotion wave:

We discovered that the model also did a very good job of adapting the locomotion wave to environmental constraints:

This suggested that the neural control system might be very useful as a control system for a snake-like robot. After a year of further development, this proved to be the case.