Jeannette Bohg

Timeline

1. 1981–2000

   From Spreewald to computer science

   Bohg is born in 1981 and grows up on a farm in Spreewald, in eastern Germany. In 2000, she finishes high school and applies to study computer science.

2. By 2012

   Degrees in Dresden and Sweden

   She earns a diploma in computer science at the Technical University of Dresden and a master’s in art and technology at Chalmers University of Technology in Gothenburg. Before her move to Max Planck in January 2012, she completes a doctorate in KTH’s Division of Robotics, Perception and Learning in Stockholm, advised by Danica Kragic. Her thesis develops methods for understanding scenes through multiple senses to help robots grasp objects.

3. 2012–2017

   Research with Stefan Schaal

   From 2012 to 2017, she works as a postdoctoral researcher and then a group leader in Stefan Schaal’s Autonomous Motion Department at the Max Planck Institute for Intelligent Systems in Tübingen. Her research combines robotic manipulation, machine learning and 3D computer vision.

4. 2017

   A lab at Stanford

   In 2017, she moves to Stanford and sets up her laboratory.

5. 2019

   Vision, touch and IEEE awards

   In 2019, she coauthors Making Sense of Vision and Touch: Self-Supervised Learning of Multimodal Representations for Contact-Rich Tasks, which wins the ICRA Best Conference Paper Award. She also receives the 2019 IEEE Robotics and Automation Society Early Academic Career Award for contributions to robot perception and learning for manipulation.

6. 2020

   Concept2Robot learns manipulation skills

   In 2020, she coauthors Concept2Robot with Lin Shao, Toki Migimatsu, Qiang Zhang and Kaiyuan Yang. The system uses human demonstration videos, reinforcement learning and imitation learning to connect language instructions to robot movements; its experiments train a policy across 78 manipulation tasks in simulation.

7. 2023

   A Sloan Research Fellowship

   She receives an Alfred P. Sloan Research Fellowship in 2023.

8. October 2023

   TidyBot learns household preferences

   In October 2023, she coauthors TidyBot: Personalized Robot Assistance with Large Language Models. The system uses large language models to learn where a person likes to store objects from a few examples, and its mobile robot puts away 85.0% of objects in real-world test scenarios.

9. January 2025

   DexForce learns from contact forces

   In January 2025, she coauthors DexForce with Claire Chen, Zhongchun Yu, Hojung Choi and Mark Cutkosky. The method measures contact forces while a person physically guides a robot, then uses those measurements to train manipulation policies; across six tasks, the policies achieve an average success rate of 76%.

10. As of October 2026

    Associate professor at Stanford

    As of October 2026, she is an associate professor of computer science at Stanford and directs the Interactive Perception and Robot Learning Lab. Her research develops perception and learning methods for autonomous robotic manipulation and grasping.