Research

Intelligent kinematic problem solving

Research Unit: 1

Project Number: 4

Example Behavior:
Individual Intelligence

Disciplines:
Behavioral Biology
Robotics

 

Doctoral Researchers:
Manuel Baum

 

Expected Project Duration
2019 - 2024


← Projects Overview

Intelligent kinematic problem solving

©Goffin Lab

Imagine that in order to escape the escape room, a robot has to solve a puzzle that consists of multiple rigid bodies and joints that can lock each other — a so-called lockbox. This is a type of sequential kinematic problem that some birds such as cockatoos can learn to solve, but as astonishing as this feat is, behavioral biology still cannot explain which characteristics empower the birds to exhibit this intelligent problem-solving behavior. In this project, behavioral biologists and roboticists get together to understand and explain this behavior in novel ways. We aim to identify prior knowledge and capabilities that the birds recruit for solving this task. To this end, we develop methodology to use robots as tools to understand animal behavior. We believe that robots are important ingredients to unveiling the mechanisms that enable cockatoos to solve lockboxes. This way, we will reveal the principles underlying kinematic problem-solving in general, such as representations or sensorimotor skills that biological and artificial systems require to solve such complex kinematic problems. We will not only better understand the biological behavior, but also improve our robotic systems. By identifying the building blocks that enable birds to solve this problem, we can either directly include these building blocks into the robotic system, or we can extract insights that will further guide robotics research. Where necessary, we will additionally develop robotic interactive perception skills along the way, thus contributing to the robotics community.


Mengers, V., Battaje, A., Baum, M., & Brock, O. (2023). Combining Motion and Appearance for Robust Probabilistic Object Segmentation in Real Time. ICRA 2023. https://doi.org/10.1109/ICRA48891.2023.10160908
Baum, M., Froessl, A., Battaje, A., & Brock, O. (2023). Estimating the Motion of Drawers From Sound. ICRA 2023. https://doi.org/10.1109/ICRA48891.2023.10161399
Baum, M., & Brock, O. (2022). “The World Is Its Own Best Model”: Robust Real-World Manipulation Through Online Behavior Selection. ICRA 2022. https://doi.org/10.1109/ICRA46639.2022.9811845
Baum, M., & Brock, O. (2021). Achieving Robustness in a Drawer Manipulation Task by using High-level Feedback instead of Planning. DGR Days. https://www.static.tu.berlin/fileadmin/www/10002220/Publications/baum21DGR.pdf

Research

An overview of our scientific work

See our Research Projects