Papers
9
Total Citations
163
H-Index
8
About
Salar Moarref is a researcher specializing in formal methods, autonomous systems, and multi-agent control, with a particular focus on bridging high-level temporal logic specifications and real-world robotic systems. His most significant contributions lie in the automated synthesis of correct-by-construction controllers for robot swarms, a challenging domain where he pioneered both top-down and compositional approaches to swarm control. Rather than relying on hand-crafted emergent behaviors, Moarref's work enables designers to specify desired collective behaviors in expressive formal languages like Linear Temporal Logic (LTL) and automatically generate decentralized controllers that provably satisfy those specifications. His 2017 paper on decentralized swarm control has accumulated 35 citations, while his complementary work on compositional and symbolic synthesis methods demonstrates the scalability of these ideas to complex multi-agent settings. Notably, his research extends beyond synthesis to skill discovery, addressing cases where a robot's action repertoire is insufficient to satisfy a given specification. With work spanning Markov decision processes, risk-averse planning, and physical robotic demonstrations, Moarref's research offers a rigorous and practically grounded framework for building verifiably safe and scalable autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3Compositional Synthesis of Reactive Controllers for Multi-agent Systems25 citations · 2016
- 4
- 5Compositional Synthesis with Parametric Reactive Controllers17 citations · 2016
- 6Verifiable Control of Robotic Swarm from High-level Specifications12 citations · 2018
- 7Finding Missing Skills for High-Level Behaviors12 citations · 2020
- 8Reactive Synthesis for Robotic Swarms11 citations · 2018
- 9Risk-Averse $ω$-regular Markov Decision Process Control2 citations · 2016