Article

OSCAR: An Adaptive Robot Navigation Framework Based on Obstacle Survival Curves

OSCAR proposes an adaptive survival-modeling framework for temporary blockages in graph-structured navigation. The robot learns residual clearance-time distributions by obstacle class from online experience and integrates them into a time-dependent graph planner with obstacle memory, computing at each blocked edge a patience threshold for how long to wait before rerouting.

Part of the event: OSCAR: A Navigation Framework that Learns Waiting Times by Obstacle

This page provides a summary and source link only.