John Qiang Gan

Text Box: Thought the harder, heart the keener.Professor of AI & Robotics

The University of Essex


Research Interests

Machine learning, artificial intelligence, robotics, signal and image processing, pattern recognition, brain-computer interfaces, human-machine interaction, intelligent systems, health informatics, data and text mining (Big Data).

Would machine learning, machine intelligence or artificial intelligence be a great driver of change over the coming decades, similar to the Internet in the past 20 years? Google and Facebook have answered this question very positively. Google paid $400m for UK artificial intelligence startup DeepMind. Facebook hired deep learning guru Prof. Yann LeCun to run its new Artificial Intelligence Lab (I visited LeCunís department at AT&T Bell Laboratories and had a discussion on the Neocognitron and his convolutional neural network over lunch in 1993).

Selected Publications

Suggested PhD Topics

(Sample work by some of my students: i++ School Newsletter, YouTube, EXPO21XX)

Selected Funded Projects


e-hpMOBE: Energy-aware High Performance Multi-objective Optimisation in Heterogeneous Computer Architecture for Biomedical Engineering Applications, funded by Spanish Ministry of Economy Competitiveness and European Regional Development Fund (collaborator).


BCI Integrated Collaborative Control of a Cognitively Enhanced Smart Wheelchair, funded by the UKIERI.


Novel Optical Paper Counting Technique, funded by the KTP.


SHOAL: Search and monitoring of Harmful contaminants, other pollutants and leaks in vessels in port using a swarm of robotic fish, funded by the Seventh Framework Programme of the European Commission.

[BBC News]


AABAC: Adaptive Asynchronous Brain-Actuated Control, funded by the EPSRC.

[Demo1: Motor imagery based Essex online BCI for simulated robot control]

[Demo2: Self-paced motor imagery based Essex online BCI for real robot control]

[Demo3: Self-paced motor imagery based Essex online BCI for simulated wheelchair control]

[Demo4: Self-paced motor imagery based Essex online BCI for real wheelchair control]

[Demo5: Self-paced motor imagery based Essex online BCI for playing hangman game]


Intelligent RoboChair, funded by The Royal Society.

[Demo1: Control by head gestures detected by a webcam]

[Demo2: Control by EMG and EOG signals detected by sensors in a headband]


Building a Fast Biped Running Robot with Biologically-inspired Neuro-mechanical Design, funded by the RPF of Essex University.

[Demo1: Biped walking robot at Essex]


Richard Gray, science correspondent from Sunday Telegraph, has tried our EMG/EOG-controlled wheelchair and brain-computer interface system. Read his articles and watch a video footage here and here.

Industrial Collaboration

Prof. Gan has developed intelligent systems and software packages for signal processing, computer vision and robotic control in the past years, in collaboration with industry in the form of contracts with companies or through the KTP programme. He is keen to collaborate with industry in the areas of sensory information processing, computer vision, robotics and intelligent control systems, data analysis and interpretation via machine learning.

Courses Taught


CE213 Artificial Intelligence (2015-2017)

CE903 Group Project (2008-2017)

CS220 Navigating the Digital World (lecturing on the AI topic, 2017- )

CE335 Digital Signal Processing (2012-2015)

CE804 Digital Signal Processing (2012-2015)

CE215 Robotics (2011-2012)

CE803 Human-Machine Interaction (2010-2011)

CE866 Computer Vision (2008-2010)

CE221 C++ Programming (2009-2010)

CC253 Programming with C++ (2007-2009)

CC481(CC461) Artificial Neural Networks (2000-2008)

CC466 Visual Guidance for Robotics (2003-2004)

CC401/406 MSc Project and Dissertation (2002-2006)

CC122 Professional Development and Practice (2001-2002)



†††††† Bridge, jogging, badminton & tennis.

[Wivenhoe Park 200 years ago and now][Click here for information about Colchester] [Back to home page]


© Copyright 1999-2016 University of Essex.
This page was last modified by John Gan in October 2016