Kegan J. Strawn
Papers
2
Total Citations
11
H-Index
2
About
Kegan J. Strawn’s research lies at the intersection of safe reinforcement learning, multi-robot coordination, and resilient autonomous systems. His work addresses two critical challenges in deploying robots in real-world, dynamic environments: ensuring safety during learning and maintaining reliable operation despite system faults. In his highly cited 2023 paper, “Conformal Predictive Safety Filter for RL Controllers in Dynamic Environments” (6 citations), Strawn introduced a novel safety filter that uses conformal prediction to provide probabilistic safety guarantees for reinforcement learning controllers navigating among unpredictable agents like pedestrians—a key step toward trustworthy robot autonomy. His 2022 work, “Byzantine Fault Tolerant Consensus for Lifelong and Online Multi-robot Pickup and Delivery” (5 citations), tackles the problem of resilient coordination, proposing a consensus mechanism that allows robot teams to continue functioning even when some robots are compromised or faulty. These contributions have practical implications for logistics, warehouse automation, and human-robot interaction. Strawn’s research is notable for bridging theoretical safety guarantees with real-time implementation, making him a rising voice in the safe autonomy community.
Research Focus
Key Achievements
Top Papers
- 1
- 2