Jianlan Luo

About

Jianlan Luo is a robotics researcher whose work sits at the intersection of reinforcement learning, robotic manipulation, and autonomous assembly, with a particular focus on bridging the gap between theoretical machine learning methods and real-world industrial applications. His research has made notable strides in enabling robots to autonomously acquire precise manipulation skills that challenge or elude conventional control approaches. Among his most influential contributions is his 2019 work on variable impedance control with reinforcement learning (177 citations), which demonstrated how integrating force/torque feedback into RL frameworks dramatically improves precision in robotic assembly. His UniGrasp system (110 citations) advanced multi-fingered robotic grasping by developing a unified model adaptable across diverse robot hand geometries. Luo has also pioneered techniques for handling deformable objects in assembly tasks (94 citations) and explored meta-reinforcement learning to accelerate adaptation to novel industrial insertion tasks (66 citations). More recently, his work on hierarchical imitation learning for cable routing and the SERL software suite reflects a commitment to making robotic RL more practical and sample-efficient. His human-in-the-loop reinforcement learning framework (2025) further underscores his drive toward dexterous, deployable robotic systems. Collectively, his research has accumulated over 670 citations, establishing him as a significant voice in modern robot learning.

Research Focus

Key Achievements

Top Papers

1. 1
   Reinforcement Learning on Variable Impedance Controller for High-Precision Robotic Assembly

   177 citations · 2019
2. 2
   UniGrasp: Learning a Unified Model to Grasp With Multifingered Robotic Hands

   110 citations · 2020
3. 3
   Deep Reinforcement Learning for Robotic Assembly of Mixed Deformable and Rigid Objects

   94 citations · 2018
4. 4
   Offline Meta-Reinforcement Learning for Industrial Insertion

   66 citations · 2022
5. 5
   Multistage Cable Routing Through Hierarchical Imitation Learning

   46 citations · 2024
6. 6
   Residual Reinforcement Learning for Robot Control

   45 citations · 2019
7. 7
   Tensegrity Robot Locomotion Under Limited Sensory Inputs via Deep Reinforcement Learning

   35 citations · 2018
8. 8
   Precise and dexterous robotic manipulation via human-in-the-loop reinforcement learning

   34 citations · 2025
9. 9
   Domain Randomization for Active Pose Estimation

   33 citations · 2019
10. 10
    SERL: A Software Suite for Sample-Efficient Robotic Reinforcement Learning

    31 citations · 2024

Key Collaborators

Contact & Links