Max Robotics — Global Service Robot Directory + US Certification

About

Zhenjun Yu is a rising researcher at the forefront of robotic manipulation, tactile sensing, and embodied AI. His work centers on enabling robots to interact with deformable objects and perform dexterous, real-world tasks through the fusion of tactile perception, deep learning, and physics simulation. Yu’s most impactful contribution to date is his 2024 paper on a deep learning-powered stretchable tactile array for capturing forceful interactions with deformable objects (24 citations), which addresses the critical challenge of occluded object geometry during manipulation—a breakthrough with direct applications in virtual reality, telemedicine, and robotics. He also developed RFUniverse, a multiphysics simulation platform for embodied AI (16 citations), providing a vital tool for training intelligent agents to handle complex, real-world physical phenomena. Further notable works include DiPGrasp, a fast, differentiable grasp planner for multi-DOF grippers (7 citations), and T-NT, a tactile perception-based method for precise robotic needle-threading using reinforcement learning. Most recently, his DexTOG framework (2024) advances task-oriented dexterous grasping guided by language, extending beyond traditional 2-finger grippers. Together, these contributions demonstrate Yu’s commitment to bridging simulation and reality, equipping robots with the perceptual and planning capabilities needed for delicate, everyday tasks.