BC3 From Brains to Robots - Thinking like an Engineer

Description

Neurorobotics is the science of the tools and the theory required to establish an interdisciplinary experimental link between robotics and neuroscience through closed-loop neural embodiment for connecting simulated brain models to robotic bodies. At the core are robots controlled by simulated nervous systems that model the structure and function of biological brains at varying levels of detail. The ability of these robots to interact with a realistic environment makes it possible to study brain models in a system where the machine brain is directly coupled to the body and where the robot’s actions influence future sensory inputs. In such closed- loop systems, the close correspondence between brain models and their neuroscientific foundation can enable a seamless exchange of knowledge between robotics and neuroscience. In this course I will give a general introduction to the field of brain-inspired robots in the first 2 sessions and will then report on the current stage of development of our Neurorobotics Platform, which is aimed at providing an integrated and integrative operational environment for scientists to develop neurorobots. The approach for the neurorobotics platform consists of providing a number of design applications for models and simulation engines that are integrated into a web-based front-end. Using this web front-end, users at different locations can rapidly construct a robot model, its brain-based controller, an environment, and an execution plan. The neurorobotics platform also allows re-use and sharing of previously defined experiments, which opens a new area of collaborative neurorobotics research, as demonstrated in the second half of the course.

Objectives

To get an overview on the emerging field of brain-inspired robotics. To learn how to use the Neurorobotics Platform for own research projects.

Literature

Material on our website neurorobotics.net

Course location

Lecture Room 4

Course requirements

none

Instructor information.

Instructor

Florian Röhrbein

Website

http://www6.in.tum.de/Main/Roehrbei