Experimental Evaluation of Road-Crossing Decisions by Autonomous Wheelchairs Against Environmental Factors

Safe road crossing by autonomous wheelchairs can be affected by several environmental factors such as adverse weather conditions influencing the accuracy of sensors based on artificial vision. Previous studies have addressed experimental evaluation of multi-sensor information fusion to support road-crossing decisions in autonomous wheelchairs. In this study, we focus on the experimental evaluation of its tracking performance against outdoor environmental factors such as fog, rain, darkness, etc. It is rather intuitive that those factors can negatively affect the tracking performance; therefore our aim is to quantify through a set of metrics how the performance of the single sensors and their information fusion changes when such external factors are present. This is a first step in designing warning strategies in a novel framework based on the MAPE-k feedback loop established for the sensor system. System reconfiguration to reduce the reputation of less accurate sensors can then be set, thus improving overall safety. The problem is analysed within the context of the European project REXASI-PRO which aims to design a trustworthy autonomous wheelchairs supported by drones in which security, safety, ethics, and explainability are entangled to improve autonomy for people with reduced mobility. Results have been achieved by using an available laboratory dataset realised for a simplified framework in a road-crossing scenario and by applying appropriate software filters to simulate different environmental conditions.