Max Yang

Max Yang

PhD Student in Robotics and AI

University of Bristol

Biography

I am a PhD student at the University of Bristol and Bristol Robotics Laboratory, under the supervision of Prof. David Barton and Prof. Nathan Lepora. Previously, I worked at Dexmate AI as a research intern. My current research focuses on sim-to-real deep reinforcement learning and multimodal perception for dexterous manipulation. My interests lie in machine learning methods for robot perception and control.

I obtained my MEng degree in Aeronautical Engineering at Imperial College London where I worked on control theory and optimal control.

Interests
  • Tactile Sensing
  • Dexterous Manipulation
  • Sim-to-Real Deep Reinforcement Learning
Education
  • PhD in Engineering Mathemtics, 2025

    University of Bristol

  • MEng in Aeronautical Engineering, 2019

    Imperial College London

Recent Publications

NeuralTouch: Neural Descriptors for Precise Sim-to-Real Tactile Robot Control
NeuralTouch: Neural Descriptors for Precise Sim-to-Real Tactile Robot Control
Text2Touch: Tactile In-Hand Manipulation with LLM-Designed Reward Functions
Text2Touch: Tactile In-Hand Manipulation with LLM-Designed Reward Functions
AnyRotate: Gravity Invariant In-Hand Object Rotation with Sim-to-Real Touch
AnyRotate: Gravity Invariant In-Hand Object Rotation with Sim-to-Real Touch
Snap-it, Tap-it, Splat-it: Tactile-Informed 3D Gaussian Splatting for Reconstructing Challenging Surfaces
Snap-it, Tap-it, Splat-it: Tactile-Informed 3D Gaussian Splatting for Reconstructing Challenging Surfaces
Bi-Touch: Bimanual Tactile Manipulation with Sim-to-Real Deep Reinforcement Learning
Bi-Touch: Bimanual Tactile Manipulation with Sim-to-Real Deep Reinforcement Learning
Sim-to-Real Model-Based and Model-Free Deep Reinforcement Learning for Tactile Pushing
Sim-to-Real Model-Based and Model-Free Deep Reinforcement Learning for Tactile Pushing
Tac-VGNN: A Voronoi Graph Neural Network for Pose-Based Tactile Servoing
Tac-VGNN: A Voronoi Graph Neural Network for Pose-Based Tactile Servoing

Experience

 
 
 
 
 
PhD in Engineering Mathematics
October 2021 – Present Bristol, United Kingdom
 
 
 
 
 
Research Intern
January 2025 – September 2025 San Francisco Bay Area (Remote), United States
 
 
 
 
 
Research and Development Engineer
September 2019 – September 2021 Cambridge, United Kingdom
 
 
 
 
 
MEng in Aeronautical Engineering
October 2015 – June 2019 London, United Kingdom
 
 
 
 
 
Research and Technology Intern
June 2018 – September 2018 Toulous, France

Projects

Visuo-Tactile Pick and Place
We present a visuo-tactile robotic system for precise food handling for the Robosoft 2023 Manipulation Competition. Our system is able to handle complex shelf configurations, with tasks including bin-picking and drink pouring.
Visuo-Tactile Pick and Place
Reinforcement Learning from Scratch
An example repository containing various reinforcement learning algorithms applied to OpenAI gym environemnts.
Reinforcement Learning from Scratch
Optimal Control for Cancer Treatment
We developed an optimal control algorithm that combines genetic algorithms and lyapunov stability theory to obtain an optimal delivery strategy of chemotherapy during the treatment of cancer.
Optimal Control for Cancer Treatment

Contact