Shushman Choudhury

Carnegie Mellon University, Stanford University

Papers

5

Total Citations

58

H-Index

4

About

Shushman Choudhury’s research lies at the intersection of robot motion planning, multi-agent coordination, and human-robot collaboration, with a focus on enabling robots to operate efficiently under uncertainty and real-world constraints. His most influential work, the Pareto Optimal Motion Planner (POMP), introduced a groundbreaking anytime algorithm that minimizes computationally expensive collision checks by searching over configuration space beliefs—a contribution that has earned 24 citations and remains a touchstone for efficient geometric path planning on roadmaps. Choudhury also tackled the complex challenge of dynamic multi-robot task allocation under time window constraints and task completion uncertainty, developing a decoupled algorithm that minimizes task failures—work cited 16 times and critical for time-sensitive applications like warehouse logistics. Further extending his impact, he contributed to multi-step mobile manipulation through a full-stack system architecture validated with real experiments, and advanced roadmap densification strategies for handling large, dense graphs. Notably, his work on incorporating qualitative human feedback into quantitative robot state estimation via Sequentially Constrained Hamiltonian Monte Carlo sampling bridges a key gap in human-robot collaboration. With a portfolio spanning theory to deployment, Choudhury’s research continues to shape how robots plan, coordinate, and interact in uncertain, dynamic environments.

Research Focus

Key Achievements

4
H-Index
5
Papers
58
Total Citations
12
Avg Citations/Paper
🏆 Most Cited Paper
Pareto-optimal search over configuration space beliefs for anytime motion planning
24 citations · 2016
📈 Most Prolific Year: 2017 (3 Papers)
🤝 Key Collaborators: 17
🏛 Institutions: Carnegie Mellon University, Stanford University

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago