Physics-Aware Conformal Prediction for Deep Learning-based Wheelchair Local Navigation

Sara Narteni, Alberto Carlevaro, Zeming Duan, Serge Autexier, Maurizio Mongelli

Abstract: When dealing with conformal prediction for real-world artificial intelligence applications, it is necessary to ensure its physical feasibility. In this work, we propose to tackle this problem for an autonomous wheelchair guided by a deep neural network for local navigation. We adapt the conformal sets to be compliant with the wheelchair’s kinematics, enhancing their efficiency while preserving coverage guarantees.

External IDs:dblp:conf/copa/NarteniCDAM25