Papers
7
Total Citations
60
H-Index
4
About
Peter Seiler is a leading researcher in the formal verification and control of complex, uncertain nonlinear systems, with a particular focus on finite-horizon reachability analysis and robust control synthesis. His most influential work, "Backward Reachability for Polynomial Systems on a Finite Horizon" (2021, 19 citations), introduces a novel optimization-based method to compute inner-approximations of backward reachable sets and synthesize admissible controllers, a critical capability for ensuring safety in autonomous systems. Building on this, his 2019 paper (14 citations) decomposes the control synthesis problem into tractable steps, enabling reliable steering of trajectories from an initial set to a target. Seiler has also pioneered the integration of formal methods with machine learning, as demonstrated in his 2020 work on tractable reinforcement learning for Signal Temporal Logic (STL) objectives (13 citations), bridging the gap between expressive task specifications and data-driven policy optimization. His recent contributions extend to trajectory-based robustness analysis and the innovative use of Large Language Models for automating control system design (ControlAgent, 2024). With a growing citation record and a focus on practical, scalable algorithms, Seiler is shaping the future of safe and intelligent control for robotics and aerospace applications.
Research Focus
Key Achievements
Top Papers
- 1Backward Reachability for Polynomial Systems on a Finite Horizon19 citations · 2021
- 2
- 3Tractable Reinforcement Learning of Signal Temporal Logic Objectives13 citations · 2020
- 4
- 5Trajectory‐based robustness analysis for nonlinear systems4 citations · 2023
- 6
- 7Backward Reachability for Polynomial Systems on A Finite Horizon2 citations · 2019