Space and nuclear robots are exposed to significant amounts of radiation which can lead to errors and system-crippling failures with great losses in economy. Moreover, limited human intervention and situational awareness necessitates that these robots are (semi)autonomous. As errors emerge, the robots should ideally be able to make their own decisions on what to do next. Moreover, they should be intelligent enough to carry out activities like drilling, grabbing, deciphering and manipulating objects on their own. All these underscore the need for a high-performance, real-time, and fault-tolerant computing platform that could serve as the brain or co-processor in a robot.
Key Challenges Being Addressed
- Customised computing for intelligent robotics
- Reliability of the computing platform
- High performance
- AI-powered real-time image processing for multi-modal robotic vision & perception
Collaborators
- University of Essex, under the National Centre for Nuclear Robotics (NCNR) hub
- University of Surrey, under the Future AI and Robotics for Space (FAIR-SPACE) hub
