Welcome to the Introbotics Lab

The Introbotics Lab is led by Prof. SiQi Zhou in the School of Computing Science at Simon Fraser University. Our vision is to build robot systems that are simultaneously performant, safe, and generalizable, paving the way for real-world autonomy in human-centric environments. We pursue this vision by drawing on the complementary strengths of machine learning and control theoretic principles to shape the robot autonomy stack, from semantic perception and contextual reasoning to reliable action.

• Where did I leave my glasses? Open-vocabulary semantic exploration in real-world semi-static environments
  IEEE Robotics and Automation Letters (RA-L), 2026
• SM²ITH: safe mobile manipulation with interactive human prediction via task-hierarchical bilevel model predictive control
  Proc. of the IEEE International Conference on Robotics and Automation (ICRA), 2026
• Semantically safe robot manipulation: from semantic scene understanding to motion safeguards
  IEEE Robotics and Automation Letters (RA-L), 2025

• L4DC'26 Tutorial on Bridging Control and Learning for Safe and Scalable Autonomy
  Organizers: Prof. Angela Schoellig (TUM), Prof. Aaron Ames (Caltech), Prof. Ryan Cosner (Tufts University), and Prof. SiQi Zhou (SFU). [webpage]
• ICRA'26 Workshop on Synthetic Data for Robot Learning
  Organizers: Prof. Fan Shi (NUS), Prof. SiQi Zhou (SFU), Dr. Mayank Mittal (ETH Zurich and NVIDIA), Dr. Ziqiu Zeng (NUS), Steve Xie (Lightwheel), Prof. Vincent Bonnet (LAAS-CNRS), and Prof. Peter Chen (UBC). [webpage]
• ICRA'26 Workshop on Semantics for Reliable Robot Autonomy
  Organizers: Prof. Angela Schoellig (TUM), Prof. Somil Bansal (Stanford), Prof. SiQi Zhou (SFU), Dr. Oier Mees (Microsoft), Lukas Brunke (TUM and UofT), Benjamin Bogenberger (TUM), Niklas Schlueter (TUM), and Haoming Zhang (TUM). [webpage]

• safe-control-gym
  A benchmark suite for safe reinforcement learning and learning-based control (913 stars, 166 forks).
• gym-pybullet-drones
  A quadrotor simulation environment for multi-agent reinforcement learning (2.1k stars, 555 forks).