Papers
6
Total Citations
55
H-Index
4
About
Chuangchuang Sun is a robotics researcher whose work bridges the critical gap between theoretical decision-making and real-world multi-agent deployment. His primary research areas span mixed-type decision-making for robotic systems, multi-agent reinforcement learning (MARL), and safety-constrained learning. Sun’s most impactful contribution is a unified formulation and nonconvex optimization method for mixed-type decision-making, which addresses the fundamental challenge of integrating continuous and discrete dynamics in tasks like task and motion planning—a paper with 17 citations. He has also made significant strides in scaling MARL by developing adaptive sparse communication graphs, enabling large-scale multi-robot coordination that avoids the exponential complexity typical of such systems. His work on automata-guided semi-decentralized MARL further advances the field by allowing heterogeneous robot teams to satisfy complex temporal logic tasks. Sun’s research on constrained meta-reinforcement learning introduces differentiable convex programming for adaptable safety guarantees, a crucial achievement for deploying learning-enabled systems in high-stakes environments. Additionally, his earlier work on designing multiaxis force/torque sensors using the force closure theorem demonstrates his versatility, offering a flexible, low-cost solution for robotic sensing. With a growing citation impact and a focus on both foundational theory and practical deployability, Sun is shaping the future of intelligent, scalable, and safe robotic systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4Automata Guided Semi-Decentralized Multi-Agent Reinforcement Learning7 citations · 2020
- 5
- 6