Sooraj Skanda
Doctoral Student @ Division of Network and Systems Engineering, KTH Royal Institute of Technology
Doctoral Student @ Division of Network and Systems Engineering, KTH Royal Institute of Technology
I am a Doctoral student at KTH Royal Institute of Technology, working at the intersection of safe reinforcement learning, multi-agent systems, and edge computing systems. My research focuses on developing cooperative algorithms that enable distributed edge agents to make intelligent, safe decisions while balancing performance, safety, and communication efficiency.
With a strong foundation in control theory and machine learning from IIT Madras, I bring together control-theoretic precision, ML adaptability, and networked system resilience to tackle real-world challenges in network systems.
KTH Royal Institute of Technology, Stockholm, Sweden
Advisors: Prof. György Dán and Prof. Viktoria Fodor
Research Focus: Cooperative and Safe Reinforcement Learning for Optimizing Edge Computing Applications
Developing algorithms that enable safe cooperation among distributed edge agents using reinforcement learning, integrating multi-agent coordination, safety enforcement, and real-time adaptation.
Indian Institute of Technology Madras
CGPA: 9.15/10 | Advisor: Dr. Rachel Kalpana Kalaimani
Thesis: Formation and Containment Control for Multi-Agent Systems with Obstacle Avoidance and Adversary Mitigation
Developed a resilient control framework using Exponential Control Barrier Functions (ECBFs) and a 2D extension of the W-MSR algorithm for adversarial resilience.
Cooperative MARL, distributed optimization, networked control systems
Control barrier functions, constrained RL, resilient control
Resource management, edge-cloud optimization, real-time systems
Reinforcement learning, game theory, cyber-physical systems
KTH Royal Institute of Technology
Developing Cooperative Safe RL algorithms for edge computing systems under network, computation, and safety constraints. Focus on multi-agent coordination, distributed decision-making, and safety enforcement using control barrier concepts.
Indian Institute of Science (IISc), Bangalore
Worked under Prof. Pavan Kumar Tallapragada on Quantized Friedkin–Johnsen dynamics, analyzing convergence and robustness in distributed decision-making processes.
Qualcomm, Noida, India
B.Tech Thesis
Developed distributed control framework for double-integrator agents with ECBFs for safety and W-MSR filtering for adversary mitigation. Theoretical guarantees for containment under attacks.
Team Amogh, IIT Madras
Developed power, communication circuits, and control logic for 5-DOF AUV. Ranked 9th globally at Singapore AUV Challenge 2024, completing sensing task in 13.71 seconds.
CFI, IIT Madras
Designed bio-inspired fish robot using silicone-based soft actuators. Won Best Project at Oceana Fest for innovation in biomimetic underwater propulsion.
Appreciation for linguistics, particularly the aesthetic aura of Kannada
Creative narrative-driven content, intellectual fiction, and cooking, fantasy shows
Stockholm, Sweden
Indian Nationality
I’m always open to discussing exciting research ideas, collaborations, or opportunities related to robotics, reinforcement learning, and multi-agent systems. Feel free to reach out — I look forward to connecting and exchanging ideas.