Sim-to-Real Transfer of Soft Robotic Navigation Strategies That Learns From the Virtual Eye-in-Hand Vision

J. Lai, T.-A. Ren (Co-first author), W. Yue, S. Su, J. Y. K. Chan, and H. Ren

• Medical scene 3D reconstruction in SOFA Framework
• Soft robot modeling and navigation with SOFTROBOTS plugin
• Closed-loop soft robot navigation with that learns from the virtual eye-in-hand vision
• Supply Video| Paper

Domain Adaptive Sim-to-Real Segmentation of Oropharyngeal Organs

Guankun Wang, Tian-Ao Ren, Jiewen Lai, Long Bai, and Hongliang Ren

• Paper

Robust, torque controlled serial 2-DoF wrist design

• Robust Wrist Teleoperation in Cabinet with Flexiv and LeapHand

Simulation of HASEL actuator via SOFA Framework

• HASEL actuator simulation

WHIPLASH: Virtual Drum Sticks with Haptic Feedback

• Stanford ME327 Final Project working with Chengyi Xing, Jinxin Li, Haotong Han
• Design a set of haptic devices that can be mounted on any sticks(apple pencil, random wood sticks found outside, etc) to serve as virtual drumsticks.
• Details: Project Wiki
• Course Website:Stanford ME327 Spring2024

Drone Perching with Deep Reinforcement Learning via Isaac Gym

• Stanford AA203 Final Project working with Yicheng Wang
• Simulator: Isaac Gym
• Reinforcement Learning for perching
• Acknowledgement: Basic simulation environment from DTU-RL-Isaac-Gym-Drone-Env
• Course Website:Stanford AA203 Spring2024

Concentric Agonist-antagonist Robots for Neurosurgical Applications

Tian-Ao Ren, Zhongchun Yu, Brian Vuong, Antonio Meola, and Allison M. Okamura

• Develop a miniature end effector equipped with two CAARs, designed to carry out surgical actions as directed by a surgeon, through its end-mounted gripper.
• RoboSoft2024: Soft Robotics Inspired Biology Workshop 2024
• Best submission for lightning talks
• Poster

Wearable for High Precision Elbow Angle Measurement

• Motion Capture System to collect ground truth data of elbow angle
• Soft Sensor System to collect sensor data
• Design experiments to verify the accuracy of mapping method

Real2Sim2Real-based Gravity-Aware Portable Soft Slender Robots

• Designed a 7-driven soft robot: 6 cable-driven actuator and 1 Sliding actuator
• Work on simulation, design, embedded coding and communication to build a Sim2Real verify platform
• Build a real machine to verify the feasibility of the platform
• IMU-Gravity-SOFA-Real-machine Performance

Floor Cruising with Dynamic Obstacle Avoidance

• Maximum speed of 2 m/s
• TEB algorithm for local planning
• Multi-sensor fusion positioning to increase localization accuracy
• Supply Video

ICRA2022 RoboMaster University Sim2Real Challenge

• Team 777 - on behalf of Beijing University of Chemical Technology (Third Prize)
• Use VINS-Mono location algorithm
• D435 + IMU mode to improve motion tracking performance and make up gaps in illumination changes
• Supply Video

Swarming Robot Navigation: Sim2Real Transfer

• Migrated multi-machine cooperative operation composite robot model in IsaacSIM
• High stability and obstacle avoidance ability
• Supply Video

Sim2Real Transfer of ARX-5s robot (a 6-DOF robotic arm)

• Supply Video