Projects
York's research and expertise is characterised by multi-disciplinary approaches to the development of advanced, ethical, safe, trusted, reliable, and certifiable mobile and autonomous robots.
Flexible cognitive robots for task completion in dynamic environments
Our aim is to develop novel human cognition inspired algorithms that can understand how humans complete tasks while dealing with variations in the environment as well as how they transfer learnt skills between various previously learnt tasks to solve new problems.
Read more about flexible cognitive robots
A Fully-Automated Robotic System for Intelligent Chemical Reaction Screening
Professor Ian Fairlamb
Automation, by intelligent chemical synthesis, is revolutionising how chemical reactions are conducted, particularly how they are optimised and characterised, improving product yields and sustainability, lowering process costs, and reducing unwanted side-products (imputrities) in final (valuable) products, e.g. agrochemical, advanced materials or pharmaceutical target compounds. The production of large chemical reaction datasets are helping considerably in this endeavour, which traditionally has been a labour-intensive process, typically with reactions being conducted one at a time. Intelligent chemical synthesis is enabling scientists to focus on data analysis and improving final product outcomes.
Autonomous Robotics Evolution: Cradle to Grave
Professor Andy Tyrrell
Imagine an environment where autonomous systems - robots - are not designed by humans, or indeed designed at all, but are created through a series of steps that follow evolutionary processes. These robots will be “born” through the use of 3D manufacturing, with novel materials and a hybridised hardware-software evolutionary architecture.
Read more about autonomous robotics evolution
Collective adaptive systems : theory and design
Collective adaptive systems consist of many autonomous units that interact in a variety of ways over multiple scales. We focus our work primarily on swarm robotic systems, developing novel approaches to self-healing systems: endowing collectives with the ability to detect, diagnose and repair failures for themselves.
Read more about collective adaptive systems: theory and design projects
Control methods for flexible and bio-inspired superlight tensegrity rovers
Dr Mark A. Post
Autonomous Robotics vehicles for planetary exploration must be responsive, energy efficient, lightweight for transport, and mechanically robust. To accomplish these goals, a small rover platform driven by the need for light weight, simplicity and reconfigurability is being developed with a minimalist mechanical structure and a flexible software design model to facilitate additions and multiple use cases. The advantages of this platform are resilience and transportability due to the low mass thin structural members of the tensegrity spine chassis, and the combined adaptability of wheeled and full-body movement. The use of flexible and distributed structures allows the rover to better adapt physically to complex and varied terrains such as those encountered on Martian slopes.
Knowledge-based self-reconfiguration for modular space robots
Dr Mark A. Post, Prof. Jim Austin
Self-configuring modular robots have the potential to revolutionize the way that tasks can be done autonomously and adaptively. One key application is in modular satellites, which can be re-configured and re-supplied with new function-specific modules to replace old or faulty ones without the risk and waste of de-orbiting the entire satellite. This project aims to extend the MOSAR (MOdular Satellite Assembly and Reconfiguration) Space Robotics SRC project with the capability, based on intelligent semantic reasoning, for a modular satellite to “know” its own capabilities and how to re-configure them to achieve science and industrial goals in orbit without human intervention.
Omni-Pi-tent modular robot platform and Dynamic Self-repair
Professor Andy Tyrrell, Professor Jon Timmis
One of the key strengths of modular robotic platforms is their ability to self-repair, ejecting failed units from their multi-robot structures and bringing in fresh modules to replace them. R.H.Peck, A.M.Tyrrell and J.Timmis's Dynamic Self-repair project seeks to pioneer ways of performing these repairs in ways which allow the robot group to maintain collective action while the ejection and replacement take place. To perform hardware experiments within this project a modular robot platform, known as Omni-Pi-tent, has been developed which provides a unique set of features. These include: an omnidirectional drive for motion across the ground to allow docking of multiple moving modules, genderless docking interfaces to let any robot attach to any other, a 2 Degree-of-Freedom hinge enabling reconfiguration in 3 dimensions, a full suite of onboard sensors to avoid dependence on external control infrastructure, and a user-friendly Raspberry Pi as each module's main computer.
Read more about Dynamic Self-repair
Self-repairing hardware paradigms based on astrocyte-neuron models (SPANNER)
Dr David Halliday, Professor Andy Tyrrell, Professor Jon Timmis
In contrast to the human brain, modern electronic systems design typically relies on a single controller or processor, which has very limited self-repair capabilities. There is a pressing need to progress beyond current approaches and look for inspiration from biology to inform electronic systems design.
This project aims to develop a new generation of self-repairing algorithms.
The social driving simulator
Dr Cade McCall
The Social Driving Simulator is designed to immerse research participants in a variety of everyday road-related situations in which they must interact with other vehicles as either drivers or pedestrians. These interactions can involve both human and autonomously controlled vehicles. As a consequence, we can study the types of interactions that naturally emerge on any shared road (from traffic jams to merging to turn-taking at ambiguous intersections) while measuring participants’ cognitive, behavioural, and physiological responses.
RoboSAPIENS:
Redefining the future of robotics trustworthy adaptation
RoboSAPIENS will develop the underlying technologies which enable robots to fully autonomously adapt their controllers and configuration settings to accommodate for unknown changes and variations while ensuring trustworthiness.
DOMINOS: Disruption Mitigation for Responsible AI
Professor Radu Calinescu
In today’s dynamic landscape, AI applications across critical sectors face continual disruptions, spanning from environmental shifts to human errors and adversities. To effectively navigate these challenges, AI solutions must demonstrate adaptability and responsibility, aligning with the diverse social, legal, ethical, empathetic, and cultural norms of stakeholders (SLEEC). Yet, current AI development frameworks fall short in addressing this multifaceted demand. DOMINOS is poised to fill this void by delivering a comprehensive methodology and toolkit for seamless development, deployment, and utilization of responsible AI solutions capable of mitigating a wide spectrum of disruptions in ways that are compliant with SLEEC norms.
BIOARCH-HS: Biomolecular tools for Archaeological, Conservation and Heritage Science
Professor Olvier Craig
The study of ancient biomolecules has revolutionised the understanding of the past and is one of the fastest growing fields in heritage science. The University of York’s BioArCh research centre stands at the forefront of this field and excels in biomolecular analysis. Despite the high demand from the heritage community, access to biomolecular analysis is often limited by capacity constraints and logistical challenges. Additionally, the facilities are currently dispersed across multiple institutions, hindering accessibility and efficiency. Robotics is going to be used to automate the pipetting of liquid for various archaeological science research. Automation builds upon positive results from three YorRobots internships.
Robotic Positioning Assistance In Breast Screening
Dr Jihong Zhu
This project aims to develop a bimanual robotic system to safely assist women during breast cancer screening, particularly focusing on helping those who have physical limitations that make traditional mammography difficult or impossible. Currently, a portion of women cannot access proper breast screening due to difficulties in achieving the correct positioning required for mammography, either because of injuries, limited upper body strength, or age-related mobility issues. The resulting technology could significantly improve access to breast cancer screening for people who currently face physical barriers. This has the potential to enable earlier cancer detection in underserved groups, ultimately contributing to better health outcomes.
Intelligent Dependable Environment Control for Sustainable Aquaculture
Dr Pengcheng Liu
Over the years, changes in the hydrological environment have introduced new issues and challenges to aquaculture and fish farming. The current operations for pond-based or sea-based aquaculture farms are highly dependent on manual labour and close human interactions with the process and cage structures. It would help if we provide fish with optimal environmental conditions by maintaining water quality, reducing stress levels, protecting against parasitic outbreaks, and ensuring there is enough food-and developing the technology to do so. The project aims to develop a dependable, cognitive, functionalized robot-assisted aquacultural platform that facilitates environment control and undertakes smart fish-farming operations such as fish feeding, fish monitoring, marine organisms collecting, water quality monitoring and analysis, net/cage cleaning, etc., regardless of the water types (freshwater, seawater) and hydrological environment (ponds, offshore).
Real-time data quality analysis and control for aquaculture prawn farming management
Dr Pengcheng Liu
Smart technologies have helped aquaculture industries in this regions to reduce labour, increase food production and boost economic growth. Studies have shown that smart aquaculture can help capture the growth, detect diseases, and classify species using machine learning approaches. However, research in smart prawn farming using computer vision are limited. The benthic nature of Macrobrachium Rosenbergii and high turbidity of the water in soil-floored ponds make it difficult for underwater cameras to capture good quality images. In this project, we investigate the use of control algorithms and machine learning techniques to optimize the data quality captured from underwater cameras to recognize and learn traits of Macrobrachium Rosenbergii and develop its growth profile.
AI-Based Real time analysis and control of the monitoring on the growth of Freshwater prawn using video image processing from underwater drone
Dr Pengcheng Liu
This project aims to increase the production of Macrobrachium rosenbergii (scampi) in Brunei by leveraging smart aquaculture technology. Current traditional methods yield low production, so the project focuses on using AI-driven solutions to enhance monitoring and management in freshwater prawn farming. It will develop an AI algorithm to improve the quality of underwater drone-captured images and create a growth model to estimate prawn age. Supported by Brunei’s fisheries department, a local SME, University of York, and funded by Brunei and UK agencies, this initiative seeks to boost prawn yields for export, benefiting Brunei's economy.
Autonomous Trajectory Learning of Robotic Arm in Contact-Rich Environments
Dr Pengcheng Liu
Real-time robot motion planning has become an active yet challenging research area recently, particularly with the issues of modelling of environmental interactions, recognition and grasping of deformable objects and optimization in contact-rich scenarios. The project aims to develop a system with real-time and closed-loop multi-sensory feedback control to realize recognition and grasping of the
unknown objects (especially deformable ones), and the inter-objects interactions can also be monitored and controlled.