EE Department Intranet -
Close window CTRL+W

ELEC70015 Human-Centered Robotics

Lecturer(s): Prof Yiannis Demiris


The aim of the module is to teach you the theory and design of robotic systems capable of intelligent interaction with humans, and their application in industrial, medical and rehabilitation settings. You will get the chance to work with research robots to develop systems capable of interacting with humans.

Learning Outcomes

Upon successful completion of this module, you will be able to: 1. build multimodal interfaces (e.g. one or more of (a) computer vision (RGB & RGB-D cameras), (b) speech recognition toolkits (c) wearable sensors (e.g. IMUs), haptic joysticks, virtual and augmented reality systems) to allow robotic systems to perceive human state information (e.g. body pose, 3d trajectories, intended action). 2. design and develop computational models of human actions (e.g. spatiotemporal sequences, or stochastic context free grammars), and use the computational models to recognize human actions using acquired human data from the multimodal interfaces above. 3. select and apply adaptive shared control methods to assist users in the control of robotic devices (e.g. the control of a robotic wheelchair to navigate safely) 4. select and apply machine learning algorithms (e.g. reinforcement learning, classification or regression) to improve the performance of the assistive robot over time.


Part A Fundamentals: Types of Human centred robotics; Contact free sharing of workspace, shared manipulation and control, cooperative assembly; Direct physical human-machine contact in medical and rehabilitation robotics; Requirements and performance metrics for human-centred robotics; Safety, flexibility, and adaptability to users. Part B Perception of user state: Interface modalities: electrophysiological signals (ExG), audio-visual signal processing, haptic interfaces; Action recognition and intention prediction using vision; Design and representation of computational models of human actions. Part C Control & Learning: Variable autonomy and adaptive shared control methods; Interactive Programming of Robots; Task learning by demonstration; Robot Learning algorithms Part D Applications: Medical and Healthcare robotics; Human-Machine Cooperative Systems in Surgery; Rehabilitation robotics; Robot Therapy following Neurological Injury; Manipulation, Mobility and Cognitive Aids.
Exam Duration: N/A
Exam contribution: 0%
Coursework contribution: 100%

Term: Spring

Closed or Open Book (end of year exam): N/A

Coursework Requirement:
         Coursework only module

Oral Exam Required (as final assessment): N/A

Prerequisite module(s): None required

Course Homepage: unavailable

Book List: