Papers
10
Total Citations
79
H-Index
5
About
Poj Tangamchit is a pioneering researcher in decentralized multirobot systems, with a career spanning foundational work in cooperative multirobot learning and behavior-based control. His key research areas include multirobot learning algorithms, task allocation, and cooperative manipulation. Tangamchit's most significant contribution is his early insight that popular single-robot learning algorithms like Q-learning, which rely on discounted rewards, fail to achieve purposeful division of labor in multirobot systems. His seminal 2003 paper, "The necessity of average rewards in cooperative multirobot learning" (30 citations), established that average-reward frameworks are essential for fostering true cooperation. He further advanced the field by demonstrating that carefully designed behavior-based architectures can enable decentralized robots to perform tightly-coupled tasks, such as the cooperative overhead transportation of a box. His work on dynamic task selection and crucial factors affecting multirobot learning has informed how researchers configure learning entities for optimal solutions. With a career spanning from 2000 to 2018, Tangamchit's research has laid critical groundwork for understanding how decentralized robots can learn and work together effectively, influencing subsequent work in swarm robotics and distributed autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1The necessity of average rewards in cooperative multirobot learning30 citations · 2003
- 2Learning-Based Task Allocation in Decentralized Multirobot Systems11 citations · 2000
- 3
- 4Dynamic Task Selection: A Simple Structure for Multirobot Systems6 citations · 2000
- 5A Comparison Study of Static and Dynamic Walking Model of A Biped Robot6 citations · 2006
- 6Crucial factors affecting cooperative multirobot learning5 citations · 2004
- 7
- 8Self-organizing approach for robot's behavior imitation4 citations · 2006
- 9
- 10