Rika Antonova
Rika Antonova researches data-efficient learning and robotics. Her career includes software engineering at Google, a robotics master’s at Carnegie Mellon, doctoral work on simulation-to-reality transfer at KTH and research at Stanford. In 2016, she coauthored an efficient optimization method for learning walking controllers in simulation. Her later publications include TidyBot for personalized robot assistance and EquiBot for adaptable robot policies. She leads the CamRAL group at Cambridge, whose research includes reinforcement learning, optimization and co-design.
- Now at
- University of Cambridge
- Works
- 47
- Citations
- 702
- h-index
- 14
Publication numbers from OpenAlex, as of 14 October 2026.
Timeline
-
2016
Learn walking controllers with fewer trials
Antonova, Akshara Rai and Christopher Atkeson developed a distance measure for Bayesian optimization of walking controllers. Their simulation experiments found policies in fewer than 100 trials across several terrain and disturbance settings. The paper presents hardware use as a possible future application.
-
As of October 2026
Study transfer from simulation
She pursued doctoral work at KTH with Danica Kragic on data-efficient simulation-to-reality transfer. She later received an NSF/CRA Computing Innovation Fellowship and researched in Stanford’s Interactive Perception and Robot Learning Lab.
-
2023–2024
Personal assistance and adaptable policies
Her publications include TidyBot, a 2023 project using language models for personalized robot assistance. She also coauthored EquiBot in 2024, which studies symmetry-aware diffusion policies for generalizable and data-efficient learning.
-
As of October 2026
Cambridge resilient autonomous learning
Antonova leads the CamRAL laboratory. The lab researches reinforcement learning, global optimization and co-design, linking algorithms with robots and their physical design.
Sources and credits
Photo: camral.github.io; creator not stated. License: Not stated by source. Source: camral.github.io. Cropped to a square and resized. Ask us to remove or replace this photo.