It is essential for robotic sensors and cameras to be reliable in the information they transmit from environments affected by radiation. It is also important for human operators to know when a sensor is failing, and how to compensate for degradation when it happens.
For this project, the Essex team tested a range of highly available sensors. They measured and modelled the effects of radiation on sensors and systems for neutron and gamma radiation. Radiation effects were measured and modelled for:
In addition, the team built machine-learning models to predict how long a sensor/robot/processor would last. They also identified the critical point for system failure. Their noise/degradation models can be fed to high-level perception algorithms.