Max Robotics — Global Service Robot Directory + US Certification

About

Kavinayan Sivakumar is a robotics and artificial intelligence researcher whose work focuses on multi-robot coordination, decentralized decision-making, and reinforcement learning in complex, uncertain environments. His most significant contributions center on applying Decentralized Partially Observable Markov Decision Processes (Dec-POMDPs) to real-world multi-robot systems, a notoriously challenging problem in autonomous systems research. Sivakumar's most influential work, "Learning for Multi-Robot Cooperation in Partially Observable Stochastic Environments with Macro-Actions" (2017), has garnered 22 citations and introduces a data-driven framework that enables teams of robots to coordinate effectively even when they lack full knowledge of their environment. By incorporating macro-actions — temporally extended action sequences — the approach bridges the gap between theoretical planning models and practical robotic deployment. Building on this foundation, his 2018 paper on adversarial policy switching addresses a critical real-world concern: making multi-agent systems robust against self-interested or hostile actors, pushing the field toward more resilient decentralized systems. Across his body of work, accumulating nearly 30 citations, Sivakumar has established himself as a thoughtful contributor to the intersection of probabilistic planning, machine learning, and cooperative robotics — areas increasingly vital as autonomous systems become more prevalent.