Department for Automation, Biocybernetics and Robotics





Projects

Author: Start year: Status: Project type: Project funding:  

Xperience (Robots bootstrapped through learning from experience)

The XPERIENCE project will implement a hybrid system, which combines, in a dynamical way, exploration based learning with the generative use of the learned experience employing structural bootstrapping to efficiently enlarge the agent’s cognitive horizon. We will build a complete robotic system for automating introspective, predictive, and interactive understanding of actions and dynamic situations and with the ability to communicate meaningfully.

More

Intellact (Intelligent observation and execution of Actions and manipulations)

IntellAct addresses the problem of understanding and exploiting the meaning (semantics) of manipulations in terms of objects, actions and their consequences for reproducing human actions with machines. IntellAct will provide means to allow for this transfer not by copying movements of the human but by transferring the human action on a semantic level.

More

Influence of ski width on alpine skiing safety
More

Learning, analysis, and detection of motion in the framework of a hierarchical compositional visual architecture

The main scientific challenge of the project lies in the design of data structures and learning processes that enable efficient learning of robust, extendable, and general­-purpose visual representations.

More

PACO+ (Perception, Action, and Cognition through Learning of Object Action Complexes)

PACO‐PLUS brings together an interdisciplinary research team to design and build cognitive robots capable of developing perceptual, behavioural and cognitive categories that can be used, communicated and shared with other humans and artificial agents. To demonstrate our approach we are building robot systems that will display increasingly advanced cognitive capabilities over the course of the programme.

More

Biologically inspired synthesis of periodic movement for a robotic humanoid leg

Control of the humanoid robot lower extremities is a complex problem and one, which has not been satisfactorily solved for tasks as bipedal walking or vertical jump. Despite the complexity of the mentioned tasks, humans can perform them with ease. Performing such task with robots demands new, alternative approaches, inspired on biological systems.

More

Robot motion synthesis through human visuo-motor learning

Development of a novel framework for robot motion synthesis that exploits human visuo-motor learning capacity.

More

Goal-directed action synthesis using a library of example movements

The aim of the project is to develop a new methodology that will enable home robots to learn new goal-directed actions in a natural way and without requiring experts to help during learning. Thus no physical models of the task will be assumed to be available because only experts can provide such models. 

More

EUROShoE (Automation of finishing operations in shoe assembly)

The goal of the project was automation of finishing operations in shoe assembly using the industrial robot.

More

Application of redundant robots in manufacturing systems

The topic of the project is the application of redundant robot in unstructured manufacturing systems which require adaptation to changing situations.

More

Control of redundant robot systems
More

ACAT (Learning and execution of Action Categories)

ACAT focuses on how artificial systems (robots) can understand and utilize information made for humans.

ACAT focuses on how artificial systems (robots) can understand and utilize information made for humans.

More