skip to primary navigationskip to content
 

Views from CAPE researchers

The Department of Computer Science and Technology worked with Jaguar Land Rover on research into Enhancing Driver Experiences through Vision Research (Endeavour) from 2015 to 2018. CAPE helped establish contact with the right people and made it easier to set up a contract. The project resulted in nine publications and a couple of invited keynote addresses. Several promising lines for further work have emerged, for which funding is still being sought.

pr 2018bProfessor Peter Robinson
Department of Computer Science and Technology, University of Cambridge

 

My research has greatly benefitted through collaboration with Jaguar Land Rover (JLR), facilitated by CAPE as one of its partners. We have developed new technologies such as active rectifiers for vehicle alternators, which improve the efficiency and power density of these devices; a new energy storage method to mitigate against the spikes in power demand in active anti-roll bar control systems; a new type of integrated starter- alternator for mild hybrid vehicles. Regular contact with JLR engineers has kept the research fresh and exciting, knowing that it is solving real difficulties in vehicle electrical power systems design. The collaboration has meant that a number of postdocs and PhD students have benefitted from industrial visits and input, which has been a great opportunity for them and their future careers. The research has led to a number of conference and journal papers, all of which have been partially credited to our main link person at JLR due to his detailed knowledge of the sector, and hence research priorities. It is this expertise, as well as the ongoing input, which has made the collaboration so valuable.

Tim flackDr Tim Flack
University Lecturer in Electrical Power and Energy Conversion

Collaboration of CAPE with BIACD

The smart in-building wireless system using flexible transmission technology (SWIFT) project is a CAPE project in collaboration with Beijing Institute of Aerospace Control Devices (BIACD). The project aims to deliver a world leading in-building wireless solution based upon Cambridge’s digital distributed antenna system (DDAS) architecture and its embedded processing algorithm.

Over the past three years, the research teams in Cambridge and Beijing have worked together to develop an industrial level DDAS prototype. This has been transformed into a series of real- life products, including a wireless distribution system that can combine 14 mobile services/ bands onto a single network infrastructure for indoor coverage. Such a system can achieve over three times higher data transmission efficiency than the current industrial standards in an optical network, and thus has lower cost and power consumption than existing solutions.

The system has the potential to carry digitised RF carriers over standard IP network infrastructure together with conventional digital data, allowing a cloud and converged network architecture to be built for multi- service and multi-operator wireless coverage in a highly efficient way. The project has also enabled business activities in both organisations collaboratively to allow DDAS products to be commercialised in both China and the UK. These include a successful patent application and a licensing agreement signed between the University and BIACD. Currently, a follow-on project is under negotiation, aiming to bring the research outcome to a new level in order to meet the new challenges of the next- generation wireless communication networks.

tongyun liDr Tongyun Li
Cambridge University Engineering Department

CAPE research with Jaguar Land Rover and Huawei

I have been working with CAPE industrial partners for six years, two years (2013-2014) with Huawei and four years (2015-2018) with Jaguar Land Rover (JLR). Projects I have been involved in are:

  • Huawei: Auto-stereoscopic mobile 3D display, developing novel methods in achieving full resolution auto-stereoscopic (glass-free) mobile display.
  • JLR: Automotive Immersive Head-Up Display (iHUD), developing next-generation head-up displays for automobiles with a large projected image plus high resolution, high brightness and depth perceptions.

Compared to fundamental research for non- industrial sponsors, research for industry is more application driven; the industrial project needs to demonstrate developed concepts. The project deliverables may also change according to the industrial partner’s needs and new mini-projects could evolve. I did not think that I could file this many patent applications (14 PCTs, 2 granted). Paper publications, however, often need to wait until the patent applications have been filed. Still, if I find research highlights during the project I am often able to publish them. I have seen direct impacts of my research; the developed demonstrator has been installed in a prototype car and was used for various tests both for our research and other research such as production teams within the industrial pIahratvneere.stablished personal contacts with the industry and we work closely with our contacts in the company. We meet as often as every two weeks and the partner company has been supportive, providing parts we need to carry out the project more efficiently.

I would highly recommend this type of experience to others within the University as it enables me to get a different perspective from industry and an appreciation for my effort. We work with company contacts as a team and it is very important for the applied work that is likely to be adopted by the industry. I am currently developing a new concept for Light Detection and Ranging (LiDAR) through the CAPE Acorn Blue Sky Research Award 2018. I plan to demonstrate my concept at the end of this year and wish to extend it into a CAPE project.

Kun Li sqDr Kun Li
Cambridge University Engineering Department

CAPE collaboration with Jaguar Land Rover

CAPE has provided us with an ideal opportunity and platform to work on industrial projects. As a researcher, this collaboration gives us a unique insight into the strategy, mindset, and tools used by world-leading technological companies. As an engineer, it is extremely exciting to be able to turn research into functional and patentable prototypes. We also get to help develop long-term research goals, steering future industry development. CAPE allows us to work with a range of brilliant professionals from a wide variety of industries and academic backgrounds, and helps us create new and innovative ideas, allowing us to get involved in high impact academic projects

>

Shreshtha Pryn aDr Matthew Pryn and Dr Pawan Shrestha
Cambridge University Engineering Department

CAPE collaboration with Disney Research

I have been working with Disney Research for three years on the project ‘Coarse Integral Hologram for 3D Images’. Through working with industrial partners, we source academic research challenges from the current non-academic projects. We can also research the non- academic project from fundamental start point. I am very happy with this type of research experience and would recommend it to other academics. In the future, I would like to continue working with Disney Research on other projects.

Jin LiDr Jin Li
Cambridge University Engineering Department

The Engineering Design Centre and the Signal Processing Group have been working together with Jaguar Land Rover (JLR) under the CAPE Partnership since 2011. During this time, it has been a facilitating channel for two PhDs, and more than five years’ worth of collaborative projects looking at key issues and technology opportunities in the automotive Human Machine Interface area. The CAPE Partnership has acted as an enabler to conceive jointly applied research projects that service the needs of JLR with a short lead-time to realisation of results.

The work has covered a wide range of research topics. One of the first and most successful has been the Motion Adaptive Touchscreen System for Automotive (MATSA), started in response to a JLR visit to the Cambridge University Engineering Department. Based on work done previously by Dr Patrick Langdon on compensation filtering for input devices for people with Cerebral Palsy, this aimed to develop a vibrating-acceleration compensation and intent recognition algorithm to be implemented as a system to improve interactive display usability in the vehicle environment. MATSA led to the development of the predictive display technology that enhances and automates the execution of the demanding task of interacting with in- vehicle systems under various road and driving conditions. It significantly improves human- machine interaction in automotive applications and enables safe, dependable control of in-vehicle systems by employing novel statistical inference techniques and suitable sensing technologies within a principled Bayesian inference framework. Experimental data from several pilot on-track studies testify to the efficacy of the developed HCI technology. It will enable JLR not only to substantially improve usability of their current in-vehicle displays, e.g. touchscreens, but also offers an effective means to easily interact with emerging display technologies, such as head-up and 3-D displays, in automotive applications via its mid-air, contactless, selection capability

Following the initial success of MATSA,two very different projects show the scope of the potential for collaboration:

The Inclusive Human Machine Interfaces (iHMI) project researched, developed and delivered a new method for developing technologies that offer opportunities to inclusively customise and personalise the in-vehicle experience. The approach was based on obtaining market research to generate a persona; carrying out ethnographic interviews with real people matching the persona. The project identified and enumerated future HMI technology likely to be usable by 2025, identifying existing inclusive design usage cases with cars in 2015. The booklets were then used as a high-quality visualisation platform for each persona usage case, presenting the same usage cases with future designs using cars with extrapolated future HMI interfaces. The main project deliverable was a boxed set of eight Doculets. A series of two internal JLR workshops on the use of the approach accompanied this, and a Techfair Display. This project was widely distributed in JLR and the method perceived as very useful. The boxed sets became a permanent resource found on most desks in R & D in Coventry.

In Driver Intent and Preferences Prediction and Profiling using Bayesian Learning from Asynchronous-heterogeneous Data (DIPBLAD), several technologies that are key to intelligent vehicles were developed and trialled using prototype implementations and subsequently patented. These developed technologies substantially benefit and aid JLR efforts to deliver a more personalised, safer and pleasant driving experience. It also enables additional JLR smart vehicle functionalities in the future especially for autonomous driving where knowing the driver-to-vehicle position (inside and outside the vehicle) and his/her profile as well as preferences can be crucial.

Through working with CAPE, a number of important generic benefits have been established.

  1. The CAPE Partnership allows the customisation of the working relationship for experimental and logistic facilities
  2. Once new technology innovations have been identified, working with JLR through CAPE greatly facilitates the patent procedure
  3. CAPE provides an “Industrialised” collaboration framework that encourages a close collaborative working environment with regular reviews and deliverables
  4. CAPE provides contact and cross-fertilisation of ideas and results between multiple projects ongoing within JLR
  5. Formulating novel solutions amenable to practical applications

 

langdonProfessor Patrick Langdon
Formerly PRA at the Engineering Design Centre, Cambridge University Engineering Department