FISHER - making waves with biologically-inspired robotics

Technology meets the natural world in one of the most innovative and creative modules of the ECS undergraduate programme.

Bringing together students from different degree programmes in ECS and different faculties across the University, the final-year Biologically-Inspired Robotics module requires students to design and build a robot, based on inspiration from the natural world.

"At present robots are very crude compared to organisms,” says Dr Klaus-Peter Zauner, joint course-leader. “This is the case at all scales from insect to elephant size, and in any category from endurance to sensing in ambiguous environments. We can still learn a lot from nature and this opens up the unusual opportunity for students to invent something significant within the scope of a single module.”

The students have just 12 weeks to research viable biological projects, devise workable solutions to the selected problem, build and demonstrate a final working device. and submit a project report.

"Our students are used to engineer to set specifications,” says Dr Zauner, “but in this module we don't set them a problem -- we ask them to come up with their own ideas. While the students find this very challenging they also regularly impress us with their innovations."

One of this year’s imaginative solutions was an innovative robotic fish, designed to imitate the natural sub-carangiform locomotion of a fish.

FISHER - a Fully Immersed Swimming Hydrodynamic Electronic Robot, used Body Caudal Fin (BCF) motion for forward propulsion, and pectoral fins for stability. The team of five ECS students - Arinze Ekwosimba, Andrew Cowan, Kenneth Payne, Jonathan Griffiths, and John Alton - designed a complex mechanical body using Computer Aided Design software, and manufactured the robot using the 3D printing and laser-cutting facilities in the ECS workshops.

A sophisticated control Printed Circuit Board was developed to mimic the segmented behaviour of an animal’s nervous system. These can be broadly categorised into the brain, communication and movement sections, and were implemented in FISHER using ARM Cortex M0 microcontrollers.

The swimming motion was facilitated by phased sine waves propagating along the fish against the direction of travel. By controlling the offset, amplitude and phase of these signals, the speed and direction of the fish could be controlled. The robotic fish was tested underwater in the University’s Lamont Towing Tank, enabling the algorithms to be optimised and the BCF motion to be comprehensively evaluated.

The project was supervised by Dr Richard Crowder, and the team benefited from support and expertise from Cambridge Circuit Company Ltd and Hammond Electronics Ltd, as well as from the state-of-the-art resources and technical staff in the ECS laboratories and workshops.

Speaking on behalf of the team, Arinze Ekwosimba summed up the experience: "The team found strength in efficiently dividing the project tasks equitably among members in a way that ensured smooth execution. Overall, we particularly enjoyed the underwater and biological challenges and were spurred to deliver something epic so as to end our degree studies on a high!"

Watch FISHER go through its paces in the Lamont Towing Tank.

"Robots have to act in real-time in a complex world and to do so with limited resources.” says Dr Zauner. “The popularity of mobile phones pushed the development of power-efficient computation and battery technology and thus created the conditions for the revolutionary developments in robotics we will certainly see in this decade."