Max Robotics — Global Service Robot Directory + US Certification

About

Junchi Liang is a robotics researcher whose work sits at the intersection of robot manipulation, machine learning, and causal reasoning. His research focuses on enabling autonomous robots to perform complex, high-precision manipulation tasks by combining computer vision, imitation learning, and structured world models. Liang's most influential contribution, "Vision-driven Compliant Manipulation for Reliable, High-Precision Assembly Tasks" (2021, 59 citations), demonstrated how state-of-the-art perception and compliant control can be unified to achieve sub-millimeter assembly precision — a longstanding challenge in robotics. This work has become a key reference for researchers tackling constrained manipulation problems. Building on this foundation, he has explored how robots can learn complex sequential tasks from just a handful of visual demonstrations, including category-level generalization across objects with varying geometries and textures using point cloud representations and dynamic graph neural networks. Liang has also made meaningful contributions to model-based reinforcement learning, developing algorithms that infer time-delayed causal relationships between events — improving both data efficiency and interpretability. With a growing citation record across manipulation, imitation learning, and causal inference, his research charts a compelling path toward robots that are simultaneously precise, adaptable, and capable of principled reasoning about their environment.