Max Robotics — Global Service Robot Directory + US Certification

About

Philip R. Osteen is a leading researcher in autonomous robotics, specializing in off-road navigation, risk-aware perception, and robot mobility in unstructured environments. His major contributions center on developing deep learning and evidential uncertainty methods to enable robots to traverse challenging terrains safely and autonomously. His most cited work, "EVORA: Deep Evidential Traversability Learning for Risk-Aware Off-Road Autonomy" (2024, 36 citations), introduces a self-supervised approach to learn terrain properties from data, automatically penalizing poor traction trajectories for faster, safer navigation. Osteen also pioneered a framework for autonomous self-righting of robots on sloped surfaces (2012, 31 citations), addressing a critical failure mode in search-and-rescue and planetary exploration missions. His research extends to semantic segmentation under out-of-distribution obstacles (2024, 9 citations), enabling robots to estimate perceptual uncertainty in novel environments. Notable achievements include his work on the RoMan human-scale mobile manipulator (2020) and integrated perception pipelines for the Robotics Collaborative Technology Alliance (RCTA). With over 90 total citations, Osteen’s work is foundational for fieldable, resilient robots operating in extreme conditions, from low-light off-road missions to dynamic disaster zones.