Bo Ai

I received my Bachelor's degree with the Highest Distinction under the Turing Program at the National University of Singapore (NUS), double majoring in Computer Science and Statistics. I work at the intersection of robotics and machine learning, with the goal of enabling robots to navigate around and interact with the real world.

Currently, I am working on learning visual-tactile dynamics models for manipulating objects with unknown physical properties at the Stanford Vision and Learning Lab (SVL), advised by Yunzhu Li and Jiajun Wu. Previously, I worked with David Hsu on control and system architecture for long-range visual navigation in diverse complex environments. Even earlier, I explored video-based action recognition under Angela Yao.

I am seeking a PhD position for the Fall 2024 intake.

Email  /  Google Scholar  /  Twitter  /  GitHub  /  LinkedIn

profile photo
Research

My overarching goal is to enable robots to assist humans in households and warehouses, which requires an integration of navigation and manipulation capabilities. To this end, I have been dedicated to building robotic systems capable of navigating large-scale, dynamic, and varied environments, as well as systems that can manipulate objects with complex dynamics.

RoboPack: Learning Tactile-Informed Dynamics Models for Dense Packing
Bo Ai*, Stephen Tian*, Haochen Shi, Yixuan Wang, Cheston Tan, Yunzhu Li, Jiajun Wu
Under review

A framework that combines visual and tactile sensing for robotic manipulation by learning a neural, tactile-informed dynamics model that can perform state estimation and dynamics prediction.

IntentionNet: Map-Lite Visual Navigation at the Kilometre Scale
Under review

A robot navigation system architecture that enables our Spot robot to navigate at kilometer scale with only crude maps and coarse localization.

Invariance is Key to Generalization: Examining the Role of Representation in Sim-to-Real Transfer for Visual Navigation
Bo Ai, Zhanxin Wu, David Hsu
International Symposium on Experimental Robotics (ISER), 2023
To appear in Springer Proceedings in Advanced Robotics (SPAR)
Paper / Book

This work examines, experimentally and theoretically, one representation that enables visual navigation policies solely trained in the indoor simulator Habitat to generalize to real-world indoor and even outdoor scenes.

Integrating Common Sense and Planning with Large Language Models for Room Tidying
Zhanxin Wu, Bo Ai, David Hsu
Robotics: Science and Systems (RSS) Workshop on Learning for Task and Motion Planning, 2023
Paper

A framework that enables an agent to put out-of-place objects back in place with only partial map information.

Deep Visual Navigation under Partial Observability
Bo Ai, Wei Gao, Vinay, David Hsu
International Conference on Robotics and Automation (ICRA), 2022
Paper / Project Page / Demo / Code

A learned controller that enables our Spot robot to navigate across varied environments under partial observations.

Whodunit? Learning to Contrast for Authorship Attribution
Bo Ai, Yuchen Wang, Yugin Tan, Samson Tan
International Joint Conference on Natural Language Processing (IJCNLP), 2022
Paper / Code

Contrastive learning improves authorship identification performance by learning author-specific representations.

Academic Service
Reviewer, RA-L, IROS 22' 23' 24', ICRA 22' 23'
Teaching
NUS CS1101S Programming Methodology, Fall 2020
NUS CS3244 Machine Learning, Fall 2022
NUS CS5478 Intelligent Robots: Algorithms and Systems (Graduate-Level), Spring 2023
Awards and Honors
2023: Dean's List (5%), NUS School of Computing
2023: Global Young Scientists Summit 2023, Singapore National Research Foundation
2022: NUS Outstanding Undergraduate Researcher Prize (~30 Recipients University-Wide, Youngest Winner), NUS
2022: NUS School of Computing Innovation Prize (2 Recipients Department-Wide), NUS School of Computing & Singapore Computer Society
2022: Certificate of Distinction in AI, NUS School of Computing
2022: Certificate of Outstanding Performance x 2, NUS School of Computing
2021: Certificate of Outstanding Performance x 1, NUS School of Computing
2018: NUS Science and Technology Scholarship (Full Scholarship, ~150K USD), Singapore Ministry of Education

"We shall not cease from exploration. And the end of all our exploring
will be to arrive where we started, and know the place for the first time." — T. S. Eliot

Template from Jon Barron. Last updated: Dec 2023.