About Me
Hello, welcome to my homepage!
I am Wenda Zhao, a Postdoctoral Researcher at Autonomous Space Robotics Lab (ASRL) working with Professor Timothy D. Barfoot. I hold a Ph.D. in Artificial Intelligence and Robotics from the University of Toronto Institute for Aerospace Studies (UTIAS), where I was advised by Professor Angela P. Schoellig. I am also an Affiliated Researcher at the Vector Institute and the University of Toronto Robotics Institute. In 2024, I was a visiting researcher at Technische Universität München (TUM) in Munich, Germany. I am also a core member of the University of Toronto Autonomous Drone Racing team, where we advanced to the finals of the first season of A2RL Autonomous Drone Championship. Before coming to Toronto, I earned a MASc degree in Electrical & Computer Engineering (ECE) from Tongji University in Shanghai, China in 2018.
My research focuses on bridging artificial intelligence and robotics technology. Meanwhile, I enjoy robotics hardware and software co-design. I am fascinated by the ability of robots to sense and interact with the physical world, which reminds us of the beauty and wonder that exist in nature.
Current Research
- State Estimation and SLAM — robust and probabilistic methods for localization, mapping, and sensor fusion in challenging real-world environments.
- Spatial AI — combining geometric localization with foundation-model-based scene understanding for robust robot navigation in complex environments.
- High-Speed Drone Racing — pushing the limits of autonomous navigation, perception, and control under extreme agility and speed.
Current Projects
- Degeneracy-Resilient LiDAR-based Teach-and-Repeat Navigation — developing lidar-based robust, topometric autonomous navigation pipelines for ground robots operating in challenging outdoor environments.
- Foundation Model-based Scene Understanding — leveraging large vision-language models (VLM) to enrich robot perception with semantic and contextual awareness beyond classical geometry.
- AI-enabled Autonomous Drone Racing — building AI-enabled drone racing system for high-speed autonomous flight in competitive racing scenarios.
Open Source Projects
- Degeneracy-resilient teach and repeat navigation using Doppler lidar. (Visual Teach & Repeat 3 (VT&R3))
- UTIL: Ultra-wideband Dataset (webpage)
- M-estimation based robust Kalman filter for UWB TDOA measurements. PR #707, (merged). Included in the Bitcraze Code Release 2021.03
- M-estimation based robust Kalman filter for UWB TWR measurements. PR #745, (merged). Included in the Bitcraze Code Release 2021.06
Contact
- Email: wenda.zhao@robotics.utias.utoronto.ca
- LinkedIn | Google Scholar | GitHub | CV (PDF)