Date: August 2019
Aurrigo, a division of the Richmond Design and Marketing Group Ltd. (RDM), announced a new Knowledge Transfer Partnership (KTP) with Aston University's College of Engineering and Physical Sciences. Lead by academics from Aston University’s Computer Science department, the partnership will develop and implement a sophisticated machine vision solution for autonomous vehicles in the Low-Speed Autonomous Transport Systems (L-SATS) sector, which will ensure that driverless cars can avoid unexpected obstacles, making their use safer.
The RDM Group, are leaders in the UK L-SATS sector and introduced the first-ever driverless vehicle in 2017. Self-driving cars is a topic increasingly gathering attention from both the research community and industry, due to its potential to radically change transport and mobility. This KTP Project Visual Perception and Decision Making for Low Speed Autonomous Transport Systems puts forward the development of intelligent systems with perception and autonomous decision capabilities in order to navigate in the presence of static and dynamic obstacles (e.g. road signs, pedestrians, cyclists, etc.) with a comprehensive understanding of the immediate environment, while following higher-level directions. The objective is to use deep-learning techniques to enable safe interaction in mixed environments to achieve full autonomous (Level 5) vehicle capability. This will require the development of an "Avoidance Dynamic Path Planning" capability, a critical innovation to achieve growth for RDM.
Controlling an agent from high-dimensional sensory inputs is a major challenge in any adaptive application. This project will exploit, for the academics, previous successes in deep reinforcement learning for traffic settings and apply it in an iterative manner between: exploration of the action space; optimisation for model update, and evaluation to verify the model’s performance.
This KTP project will adopt a staged approach with the first stage utilising a simulated environment for training and testing. This will enhance the team's understanding, enable the design of suitable model architectures and hyperparameter selection. Once the navigation task using topological maps to represent the 3D world in the simulated environment is achieved, the project will proceed to the second stage, involving on-the-fly visual detection and tracking of obstacles using deep learning and transferring learning from the simulated task to the real-world application.
The Academic team for this KTP includes Dr George Vogiatzis and Dr Diego Faria. Dr George Vogiatzis has extensive research in the subjects of Machine Vision, Machine Learning and Artificial Intelligence; while working at Toshiba, he evaluated obstacle avoidance strategies and developed algorithms to improve upon them. Dr Diego Faria has collaborated on multiple EU and international projects within the fields of Artificial Perception, Cognitive Robotics including Human-Robot Interaction and Autonomous Systems. Both academics have extensive experience of working collaboratively with business to transfer knowledge and create practical solutions.
Nick Ridler, engineering manager at RDM Group, says:
“The expertise of Aston’s academics and their specialist labs at Aston University are key, allowing us to run detection and recognition of obstacle scenarios. This data will be used in our decision-making algorithms which will choose the optimal solution based on the environment. This project will allow us to run accurate detection and obstacle recognition scenarios without having to 'field test' our driverless vehicles.”
Further benefits of the KTP include:
- Higher levels of safety in operating autonomous vehicles
- Replacement of high-cost detection Systems, e.g. LIDAR, by low-cost alternatives, improving RDM Group’s competitiveness
- Delivery of new Low-Speed Autonomous Transport Systems (L-SATS) into mixed environments.
Expected project finish date: September 2022
This project is co-funded by the UK’s innovation agency, Innovate UK.
Knowledge Transfer Partnerships is a UK-wide programme that has been helping businesses for the past 45 years to improve their competitiveness and productivity through the better use of knowledge, technology and skills that reside within the UK Knowledge Base.
To find out more about how your business can benefit from working with Aston University, please email firstname.lastname@example.org or call 0121 204 4242
Date: November 2020
James Heaton takes on the role as KTP associate
After a thorough recruitment process to ensure the right person was hired to drive this Knowledge Transfer Partnership to its full potential, in September James Heaton was recruited as the KTP associate. The objective of this collaboration is to enable Aurrigo vehicles to run at Level 5 autonomy, which means they will run outside confined vehicle free spaces, e.g. airports and closed campuses. This will require the development of an "Avoidance Dynamic Path Planning" capability, a critical innovation to achieve growth for RDM.
We caught up with James to discuss this exciting, innovative project.
1. What attracted you to the role and how did you feel about being recruited onto the project?
I went for this role as I felt that it was a great position to apply my skill set and education; I did my Bachelor’s degree in Mechanical Engineering with a final year specialisation in robotics, and I recently completed my master’s in Artificial Intelligence. Naturally, working with autonomous vehicles was a perfect fit. I am very pleased about being recruited onto the project. I got on with everyone on the team instantly and settled into my role very quickly.
2. What have you been working on in your first two months?
In my first couple of months I have been working with a Far East Airport and Regulators to create a simulation environment for their airport. 3D laser scans were taken of the runway and surrounding buildings which I have turned into a rendered world using our processing pipeline linked to a game development platform. Using CAD models of the vehicles, we can place them in the world and have them drive around using a sophisticated physics engine. This allows us to test the running of these autonomous vehicles in the simulation without the need to be physically present at the airside. This same process can be used for any location where we can either record or be provided with a laser scan. We are also working on a projects at other European airports.
3. What are you looking forward to the most, as you progress through this project?
Aurrigo has a number of interesting applications for their autonomous vehicles ranging from small pods to high-capacity shuttles used for public transport. However, what I am most excited to work on is the autonomous baggage handling in airports. Whilst autonomous vehicles are constantly reported on in today’s media, these are almost always self-driving cars on public roads, and not much is known to the public about how we can use artificial intelligence and autonomous vehicles to improve the efficiency of baggage handling (even I had not considered this as a potential application until my interview!). I am relishing the technical challenges of this application; most research into autonomous vehicles has been constrained to public roads, but the nature of the environment changes drastically in an airport where there are unique vehicles you would not see on the road (such as aircraft steps, baggage trailers etc.), as well as the aircraft themselves.
We look forward to following James' work as the project progresses over the next couple of years.