Thomas J. Collins
Papers
5
Total Citations
32
H-Index
4
About
Thomas J. Collins is a leading researcher in self-reconfigurable modular robotics, with a focus on enabling autonomous systems to adapt, locomote, and manipulate objects in complex environments. His major contributions include the development of **ReMod3D**, a high-performance physics-based simulator that allows researchers to safely and cost-effectively model the behavior of large-scale modular robot swarms. Collins also pioneered a **near-optimal dynamic power sharing scheme** that significantly extends the operational lifespan of interconnected modules, a critical step toward practical, long-duration deployments. His work on **autonomous 6D-docking and non-stationary-base manipulation** demonstrates how modular robots can self-aggregate to achieve fault-tolerant locomotion and dexterous manipulation. More recently, Collins has explored **autonomous learning of POMDP state representations from surprises**, pushing the boundaries of how robots can adapt to partially observable, stochastic environments without human-engineered features. With cumulative citations exceeding 30 across his top papers, Collins’ research is foundational for the next generation of resilient, self-organizing robotic systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5Autonomous learning of POMDP state representations from surprises2 citations · 2018