Thushal Wijekoon Mudiyanselage
Papers
1
Total Citations
39
H-Index
1
About
Thushal Wijekoon Mudiyanselage is a robotics researcher whose work lies at the intersection of reinforcement learning, bipedal locomotion, and adaptive control. His most cited contribution, "Balance control of a biped robot on a rotating platform based on efficient reinforcement learning" (2019, 39 citations), introduces a novel hybrid framework that merges model-based and model-free reinforcement learning to stabilize a NAO robot on a rotating platform with unknown angular velocity. This approach treats the platform’s motion as an external disturbance, enabling the robot to maintain balance without prior knowledge of the disturbance dynamics—a significant step toward robust, real-world legged locomotion. By demonstrating that a learning agent can adapt to unpredictable environmental forces, Wijekoon Mudiyanselage’s work has implications for disaster response, humanoid robotics, and autonomous systems operating in dynamic settings. His research highlights the power of combining data-driven learning with physical models, offering a pathway to more resilient and sample-efficient control strategies. For students and researchers in robotics and AI, his work exemplifies how blending theoretical reinforcement learning with practical hardware challenges can push the boundaries of autonomous mobility.
Research Focus
Key Achievements
Top Papers
- 1