Tianle Huang
Papers
2
Total Citations
11
H-Index
2
About
Tianle Huang is a robotics researcher advancing sim-to-real transfer for soft and lightweight robotic systems. His primary research areas include domain randomization, system identification, and reinforcement learning for buoyancy-assisted legged robots. Huang’s most notable contribution is the development of BALLU (Buoyancy Assisted Lightweight Legged Unit) robots, which leverage unique soft and light characteristics to enable intrinsically safe human-robot interaction—a stark contrast to traditional heavy, rigid robots. In his 2023 work “Residual Physics Learning and System Identification for Sim-to-real Transfer of Policies on Buoyancy Assisted Legged Robots” (9 citations), he pioneered a method combining residual physics learning with system identification to overcome the challenges of modeling BALLU’s sensitive dynamics. His 2024 paper “BayRnTune: Adaptive Bayesian Domain Randomization via Strategic Fine-tuning” (2 citations) introduces an adaptive approach to domain randomization, automating the tuning of randomization parameters that previously required extensive manual effort. This work builds on Bayesian Domain Randomization principles, offering a more efficient path from simulation to real-world deployment. Huang’s research is particularly impactful for developing robots that can safely operate alongside humans in domestic and healthcare settings, where traditional rigid robots pose risks.
Research Focus
Key Achievements
Top Papers
- 1
- 2