Papers
21
Total Citations
247
H-Index
8
About
Bo Song is a robotics and artificial intelligence researcher whose work spans rehabilitation robotics, humanoid systems, robot learning, and vision-based manipulation. His most influential contribution, "Human-in-the-Loop Control Strategy of Unilateral Exoskeleton Robots for Gait Rehabilitation" (2019, 48 citations), introduced an innovative methodology for assisting hemiplegic patients through coordinated unilateral exoskeleton systems, demonstrating his commitment to socially impactful robotics. Song has also made significant strides in reinforcement learning, developing novel reward-shaping techniques such as Relay Hindsight Experience Replay (38 citations) and Dense2Sparse reward frameworks that balance learning efficiency with effectiveness in robot manipulation tasks. His work on humanoid wheeled robots employing fuzzy impedance control (27 citations) reflects a broader interest in adaptive, human-like robotic behavior. Song further contributes to programming-by-demonstration, vision-based manipulation without camera calibration, and sports robotics — notably advancing table tennis robot systems capable of handling high-speed spinning balls. Collectively accumulating over 200 citations, his research consistently bridges theoretical machine learning with real-world robotic applications, making him a notable voice in intelligent and interactive robotics systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3Whole-Body Fuzzy Based Impedance Control of a Humanoid Wheeled Robot27 citations · 2022
- 4
- 5A Novel Trajectory-Based Ball Spin Estimation Method for Table Tennis Robot16 citations · 2023
- 6
- 7Visual servoing using non-vector space control theory11 citations · 2012
- 8
- 9
- 10D2SR: Transferring Dense Reward Function to Sparse by Network Resetting8 citations · 2023