Automotive grade printed and flexible optical bus for self-learning and autonomous cars
Lead Research Organisation:
University of Cambridge
Department Name: Engineering
Abstract
Title:
Investigation of Optical Couplers for the Application in Optical Bus Systems for Autonomous Cars
Summary:
Challenges and the aim of the project:
JLR have tried optical networks in MOST networks within the infotainment area. They did not deliver the best performance as there were some inherent flaws in the implementation, especially at transducing end. CE (Consumer Electronics) spec of optical networks seems to have solved that problem, and faster networks with more bandwidth are now being implemented industry wide from home broadband to industrial communications. Optical network is truly future proof and can support really high bandwidths for coming few decades. They are extremely light and with new developments can withstand higher temperatures. Can automotive industry look at it again in new light to realize the benefits of OFC (Optical fibre communications) for future cars?
On the roadmap to autonomous cars, the in-car electrical architecture will experience an overhaul. More autonomous features rely on more sensors in and around the car. Currently as it stands premium cars use about 70+ ECUs (Electronic control units) and 110+ sensors in the car. More will be added as we move towards self-driving car. High definition camera sensors for assistance and high definition displays and ubiquities connectivity will be taken for granted in such premium cars. In our estimation our data buses and in-car network shall be able to support data rates between 10Gbps to 50 Gbps depending on the configurations. Our current in-car electrical architecture and network quality/speeds are totally inefficient for such a vision. Traditional copper based network is also inefficient and too heavy to meet future vehicle requirements our premium cars today have about 6+ Km of copper that weighs between 60 to 90 kilos depending upon the configuration. With current estimations it's highly likely that a distributed computing architecture with a fast, high bandwidth and reliable network as a backbone will be needed for future cars. Optical waveguides have proven to carry high speed data reliably and they are really light weight, combined with printed and flexible electronics can prove one of the solution to this foreseeable future challenge. Polymer optical waveguides have attracted more interests in the area of optical communication networks in autonomous cars since they are flexible and can be formed by printing techniques, which means they can be easily integrated with already existing circuits. Apart from the waveguides, optical couplers are also of vital importance in a complete optical system because they determine the power efficiency and the signal quality of the whole system. Therefore, the aim of this PhD project is to investigate the proper design of optical couplers for the optical system in autonomous cars.
Scope: Following work shall be included in the scope for this PhD.
Investigation of different coupling methods
Design of couplers for polymer optical waveguides for optical networks in cars
Fabrication and characterisation of designed couplers
Investigation and design of couplers with tuneable coupling ratio
Testing the system to automotive requirements, specification and failure modes
Network simulation to create test scenarios and physical layer simulation is also required for further studies and tests.
Outcomes:
A prototype of a system with polymer optical waveguides and couplers delivering data between sensors and ECUs in a preferred topology.
A PhD thesis into design of optical couplers for polymer optical waveguides in automotive domain containing challenges, manufacturing direction, failure modes and recommendations.
Investigation of Optical Couplers for the Application in Optical Bus Systems for Autonomous Cars
Summary:
Challenges and the aim of the project:
JLR have tried optical networks in MOST networks within the infotainment area. They did not deliver the best performance as there were some inherent flaws in the implementation, especially at transducing end. CE (Consumer Electronics) spec of optical networks seems to have solved that problem, and faster networks with more bandwidth are now being implemented industry wide from home broadband to industrial communications. Optical network is truly future proof and can support really high bandwidths for coming few decades. They are extremely light and with new developments can withstand higher temperatures. Can automotive industry look at it again in new light to realize the benefits of OFC (Optical fibre communications) for future cars?
On the roadmap to autonomous cars, the in-car electrical architecture will experience an overhaul. More autonomous features rely on more sensors in and around the car. Currently as it stands premium cars use about 70+ ECUs (Electronic control units) and 110+ sensors in the car. More will be added as we move towards self-driving car. High definition camera sensors for assistance and high definition displays and ubiquities connectivity will be taken for granted in such premium cars. In our estimation our data buses and in-car network shall be able to support data rates between 10Gbps to 50 Gbps depending on the configurations. Our current in-car electrical architecture and network quality/speeds are totally inefficient for such a vision. Traditional copper based network is also inefficient and too heavy to meet future vehicle requirements our premium cars today have about 6+ Km of copper that weighs between 60 to 90 kilos depending upon the configuration. With current estimations it's highly likely that a distributed computing architecture with a fast, high bandwidth and reliable network as a backbone will be needed for future cars. Optical waveguides have proven to carry high speed data reliably and they are really light weight, combined with printed and flexible electronics can prove one of the solution to this foreseeable future challenge. Polymer optical waveguides have attracted more interests in the area of optical communication networks in autonomous cars since they are flexible and can be formed by printing techniques, which means they can be easily integrated with already existing circuits. Apart from the waveguides, optical couplers are also of vital importance in a complete optical system because they determine the power efficiency and the signal quality of the whole system. Therefore, the aim of this PhD project is to investigate the proper design of optical couplers for the optical system in autonomous cars.
Scope: Following work shall be included in the scope for this PhD.
Investigation of different coupling methods
Design of couplers for polymer optical waveguides for optical networks in cars
Fabrication and characterisation of designed couplers
Investigation and design of couplers with tuneable coupling ratio
Testing the system to automotive requirements, specification and failure modes
Network simulation to create test scenarios and physical layer simulation is also required for further studies and tests.
Outcomes:
A prototype of a system with polymer optical waveguides and couplers delivering data between sensors and ECUs in a preferred topology.
A PhD thesis into design of optical couplers for polymer optical waveguides in automotive domain containing challenges, manufacturing direction, failure modes and recommendations.
People |
ORCID iD |
| Daping Chu (Primary Supervisor) | |
| Peng Wang (Student) |
Studentship Projects
| Project Reference | Relationship | Related To | Start | End | Student Name |
|---|---|---|---|---|---|
| EP/N509103/1 | 01/10/2015 | 31/03/2022 | |||
| 1800898 | Studentship | EP/N509103/1 | 01/10/2016 | 31/03/2021 | Peng Wang |
| Description | The performance of the polymer waveguides under different conditions, for example when bending or twisting is applied on them, needs to be investigated in order to examine the robustness of the system. Apart from the investigation on bending and twisting performance that had been carried out by other researchers, in my experiment the performance of the waveguide was investigated when it is stretched by different amount of force before it reaches the plastic deformation region. It was found that the change of the loss of the waveguide was negligible. This means the performance of the waveguide is stable under certain amount of tension. Apart from the performance of the waveguide itself, the coupling between waveguides is also a key requirement inside an optical bus system since the signal always need to be communicated between branches and the bus. Therefore, different ways of coupling light into and out of from a polymer waveguide were investigated. Directional coupling in which two waveguides are brought close to each other to get light coupled from one to another was carried out on the samples we received from Dow Corning. It was found that the coupling efficiency was extremely low. The reason for this is that the surface roughness of the sample makes the gap between the waveguides to be much larger than the evanescent field extension which makes it not satisfying the key requirement of the directional coupling method. In light of this, an index matching method is now under investigation, in which items like index matching gel will be applied on the surface of the waveguide to let light coming out and collected by another one. Other coupling methods such as grating coupling is also now being investigated by doing simulations and experiments. |
| Exploitation Route | The findings of this research could be taken by industries as a reference of making flexible optical systems that could be used in variety of areas that needs fast, short range and low cost communication systems, such as autonomous cars, air planes, data centers, etc. |
| Sectors | Digital/Communication/Information Technologies (including Software),Transport,Other |