Design of a Multi-Sensing Platform for Collaborative and Safe Operation of Autonomous Systems

The operation of autonomous systems is prone to many problems related to safety and security. The costs of most of the sensors and actuators used in autonomous systems are very high. Furthermore, the sensor hardware and software could malfunction during operation due to adverse weather conditions or cyber-attacks. The safe and secure operation of autonomous systems and the state of health (SoH) of the sensors used in such systems require in-depth research and experimental evaluation. The use of different types of sensors in autonomous systems also necessitates new methods that process the data from these sensors, give weighting for the data from each sensor, and verify the outputs that could determine the position, speed, and angular orientation of each object in the autonomous system and any obstacles in the environment around it. For that purpose, this paper proposes a multi-sensing platform. The platform identifies sensor malfunctions and restores the normal system operation for a robotic arm and an autonomous vehicle. Various methods of interfacing ten different types of sensors to each object are presented to increase the accuracy and fault tolerance of the detection of obstacles using the multi-sensor platform. The results indicate that the proposed method could potentially avoid risks due to a single or multiple sensor malfunction for various systems.