Max Robotics — Global Service Robot Directory + US Certification

About

Joshua Joseph’s research lies at the intersection of safe autonomous navigation, machine learning, and language grounding, with a focus on enabling robots to operate reliably in complex, uncertain environments. His most influential work, “Probabilistically safe motion planning to avoid dynamic obstacles with uncertain motion patterns” (287 citations), introduced a real-time path planning algorithm that guarantees probabilistic feasibility for robots facing process noise and unpredictable obstacles—a critical advance for deploying autonomous systems in crowded, real-world settings. Joseph further explored how robots can learn from imperfect models in “Reinforcement learning with misspecified model classes” (21 citations), addressing the challenge of acting when true world dynamics are unknown. His work on “Learning perceptually grounded word meanings from unaligned parallel data” (59 citations) bridges robotics and natural language processing, enabling machines to acquire semantic understanding without explicit alignment. Earlier contributions include nonparametric Bayesian behavior modeling for dynamic environments. Through these contributions, Joseph has advanced both the theoretical foundations and practical safety of autonomous systems, making his research essential reading for those working on robust, learning-enabled robotics.