Swarthmore College Department of Computer Science

Lisa Meeden: Developmental Robotics

robot picture

Developmental robotics is a relatively new approach that focuses on the autonomous self-organization of general-purpose, task-independent robot control systems. It takes its inspiration from developmental psychology and developmental neuroscience. Developmental robotics is a move away from the engineering approach where a robot is designed to solve a particular pre-defined task (such as path planning to a goal location). In contrast, developmental robotics explores the kinds of perceptual, cognitive, and behavioral capabilities that a robot can discover through self-motivated actions based on its own physical morphology and the dynamic structure of its environment.

In particular, I am interested in how a robot could be designed so as to choose actions based on self-motivation. My collaborators and I have begun to build such an architecture using a connectionist framework. In our model, self-motivation is viewed as an emergent property arising from two competing pressures: the need to accurately predict the environment while simultaneously wanting to seek out novelty in the environment. The robot's internal prediction error about how its sensory input will change over time is used to generate a reinforcement signal that pushes the robot to focus on areas of high error or novelty.

Research papers: http://www.cs.swarthmore.edu/~meeden/#research