Joanne Truong
Papers
8
Total Citations
178
H-Index
8
About
Joanne Truong is a leading researcher at the intersection of embodied AI and robotics, whose work fundamentally rethinks how robots learn to navigate and interact with the real world. Her central research areas include sim-to-real transfer, legged locomotion, and mobile manipulation. Truong’s most impactful contribution is her critical examination of the sim-to-real gap, most notably in her highly cited work “Are We Making Real Progress in Simulated Environments?” (39 citations), where she developed tools to measure whether simulation progress translates to real-world robot performance. She has pioneered approaches like ViNL (28 citations), which enables quadrupedal robots to step over obstacles in unseen environments, and ASC (36 citations), which coordinates basic skills for long-horizon mobile manipulation tasks. Her innovative “Rethinking Sim2Real” paper (13 citations) challenges conventional wisdom by showing that lower-fidelity simulation can actually improve real-world transfer. Truong has also achieved remarkable zero-shot transfer results, including navigating outdoors without any outdoor training data (I2O, 9 citations). Her work consistently bridges simulation and reality, making her a key figure in advancing practical, deployable robotic systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2ASC: Adaptive Skill Coordination for Robotic Mobile Manipulation36 citations · 2023
- 3ViNL: Visual Navigation and Locomotion Over Obstacles28 citations · 2023
- 4Learning Navigation Skills for Legged Robots with Learned Robot Embeddings21 citations · 2021
- 5Sim-to-Real Transfer for Vision-and-Language Navigation21 citations · 2020
- 6
- 7
- 8