SWARM

Self-organising Wide area Autonomous vehicle Real-time Marshalling

Autonomous vehicles

Autonomous, or self-driving, vehicles have been hard to miss in the news recently, whether this is Tesla's partially automated 'Auto Pilot' feature, or the fully driverless 'pods' that arrived on the streets of Milton Keynes in October 2016.

As the technology becomes more familiar, people are becoming increasingly confident that individual vehicles will be able to drive and navigate themselves on roads and around people. But a single self-driving car is of limited value, it needs to work as part of an existing transport system – therefore conversations are now moving towards ‘how will they actually work in a city network’. Currently, one pod can move one person (maybe two if sharing), but it needs a trained safety driver to be in the vehicle, plus traffic cameras to monitor its every move and to make sure it does what is expected.

This is expensive, so to make autonomous urban transport more efficient - while maintaining safety - we need to share this supervision between the pods and external systems (cameras and humans). We aim to achieve this by using Swarm Intelligence (what bees or ants do when part of a colony) to enable real-time, collaborative supervision of pods – meaning individual pods are locally supervised, not only by cameras or humans but by neighbouring pods in the Swarm colony.

Multi-objective Optimisation of Pod Fleet

Technology inspired by ‘Mother Nature’ is being developed to help manage fleets of driverless pods. The concept is based on fusing together existing information from other vehicles in the fleet to allow each pod to locally decide the most appropriate action for the group as a whole – similar to how insects and birds currently behave. This means that pods can highlight any unexpected behaviour to a supervisor, as well as giving local authorities the chance to take advantage of ‘platooning’, where vehicles follow each other when possible to minimise the number or individual vehicle movements. The technology also makes the system automatically adapt its behaviour to meet demand so that pods can be optimally distributed within a city to the areas where they are most likely.

Co-simulation of Autonomous Vehicles

This project will use WMG’s ‘3xD simulator for Intelligent Vehicles’ in order to conduct remote co-simulations between a simulated pod and a physical pod, located at RDMs Urban Development Lab).

Publications

• Kampert, Erik, Schettler, Christoph, Woodman, Roger, Jennings, Paul A., Higgins, Matthew D., 2020. Millimeter-wave communication for a last-mile autonomous transport vehicle. IEEE Access, 8, pp. 8386-8392. DOI: 10.1109/ACCESS.2020.2965003
• Woodman, Roger, Lu, Ke, Higgins, Matthew D., Brewerton, Simon, Jennings, Paul A., Birrell, Stewart A., 2019. Gap acceptance study of pedestrians crossing between platooning autonomous vehicles in a virtual environment. Transportation Research Part F: Traffic Psychology and Behaviour, 67, pp. 1-14. DOI: 10.1016/j.trf.2019.09.017
• Woodman, Roger, Hill, William D., Birrell, Stewart A., Higgins, Matthew D., 2019. An evolutionary approach to the optimisation of autonomous pod distribution for application in an urban transportation service. 2019 23rd International Conference on Mechatronics Technology (ICMT), Salerno, Italy, 23-26 Oct 2019, Published in 2019 23rd International Conference on Mechatronics Technology (ICMT), pp. 1-6. DOI: 10.1109/ICMECT.2019.8932138
• Woodman, Roger, Lu, Ke, Higgins, Matthew D., Brewerton, Simon, Jennings, Paul. A., Birrell, Stewart A., 2019. A human factors approach to defining requirements for low-speed autonomous vehicles to enable intelligent platooning. 2019 IEEE Intelligent Vehicles Symposium (IV), 9-12 Jun 2019, Paris, France, pp. 2371-2376. DOI: 10.1109/IVS.2019.8814128

RG4CAV

Robust GNSS for Intelligent Vehicles

Research into Connected and Autonomous Vehicles (CAVs) is currently seen as of high national importance. A key component and technological critical path to their ultimate deployment is the provision of robust navigation systems.

PhD Studentship funded by Spirent Communications Ltd with support from WCPRS

The main challenge with (CAVs) is related to the difficulties in obtaining precise localisation and timing. While techniques such as object detection and sensor data fusion have been adopted, the location errors reported are below the specified intelligent transportation systems (ITS) requirements. In applications such as cooperative positioning precise timing is required for range-based vehicle and infrastructure localization. With respect to positioning accurate location information is required for lane detection, manoeuvring on a winding and collision avoidance.

The RG4IV project focuses on the robustness of global navigation satellite systems (GNSS) from the perspective of intelligent vehicles. Robustness, from a CAV point of view in this project, is seen as being able to withstand both physical and cyber challenges. Thus, this project is holistic because the RF signal propagation characteristics, which are dependent on weather, location, and other deployment scenario constraints, will be considered alongside the associated security aspects of the system.

This project is funded by Spirent Communication Ltd. and WCPRS. The IV connectivity team's role involves conducting measurements to test GNSS signals using a GSS 7000 signal generator provided by Spirent. The multi-GNSS constellation Simulator System will allow the researchers jointly with the WMG's 3xD simulator to depict real-life RF signals conditions for Autonomous Vehicle scenarios. WMG will contribute towards providing innovative solutions to the GNSS robustness which will subsequently aid multiple leveraging technologies.

5G∀IV

Connected and autonomous vehicles (CAVs)

Connected and autonomous vehicles (CAVs) will depend on fast and reliable vehicle-to-vehicle (V2V) and vehicle-to-everything (V2X) telecommunication, as they have to rely not only on information from their own sensors, but also on that of other vehicles and their environment.

5G

This poses significant challenges to the underlying communication system, as information must reliably reach its destination within a short period, beyond what current wireless technologies can provide. In addition, the high data rates necessary for the exchange of massive amounts of raw sensor data are not yet supported by current technologies either.

5G, the fifth generation of mobile communication technology, holds the promise of improved performance in terms of reduced latency, increased reliability, and higher throughput under higher mobility and connection densities. Moreover, the use of millimetre waves (mmWave) as carriers enables access to the required larger bandwidths. The biggest challenge for the use of mmWave carrier frequencies (CFs) is the high propagation path loss and susceptibility to blockage by building materials. Hence, before implementing mmWave technology into vehicles, it is crucial to understand such large-scale fading parameters and be able to represent the channel conditions accurately in realistic channel models. Complementary to this, multipath components (MPCs) due to in-vehicle mmWave reflections need to be further understood and described in terms of small-scale fading.

Knowledge about the channel conditions is crucial for increasing the efficiency of multiple access techniques, with channel estimation, beamforming and handover depending critically on the channel dynamics. To achieve this level of understanding, detailed vehicular channel measurements need to be performed, covering realistic scenarios and RF technology deployment, and addressing industrial and academic challenges.

Faraday-shielded anechoic chambers

The team’s channel-sounding equipment in the 3xD Simulator and NAIC Communications and Sensors Lab is used for 5G V2X communication measurements covering all vehicle-specific scenarios. Moreover, the University of Warwick campus provides the opportunity to extend this work to outdoor, urban scenarios. The results of our 5G channel-sounding campaigns are used to develop and validate a 5G mmWave V2X channel model for frequencies between 18 and 85 GHz. This knowledge is subsequently used in our stochastic geometry-based modelling work related to mmWave coverage and channel capacity in vehicular scenarios. Both our measurement and modelling results will be disseminated in leading journals and on international flagship conferences, thereby impacting on the academic community, whilst CAV developers can use the knowledge to improve and accelerate their radio and sensor design from a telecommunication perspective.

Publications and Press Releases

[Journal]: "Millimeter-Wave Communication for a Last-Mile Autonomous Transport Vehicle" in IEEE Access [DOILink opens in a new windowLink opens in a new windowLink opens in a new window / UoW OA Research RepositoryLink opens in a new windowLink opens in a new windowLink opens in a new window]

[Journal]: "Connectivity Analysis for mmWave V2V Networks: Exploring Critical Distance and Beam Misalignment" in 2019 IEEE Global Communications Conference (GLOBECOM) [DOILink opens in a new windowLink opens in a new windowLink opens in a new window / UoW OA Research RepositoryLink opens in a new windowLink opens in a new windowLink opens in a new window]

[Press Release]: WMG has set an autonomous vehicle communications record using 5G [linkLink opens in a new windowLink opens in a new windowLink opens in a new window]

[Journal]: "A grid-based coverage analysis of urban mmWave vehicular ad hoc networks" in IEEE Communications Letters [DOILink opens in a new windowLink opens in a new windowLink opens in a new window / UoW OA Research RepositoryLink opens in a new windowLink opens in a new windowLink opens in a new window]

[Journal]: "Investigating the V2V Millimeter-Wave Channel near a Vehicular Headlight in an Engine Bay" in IEEE Communications Letters [DOILink opens in a new windowLink opens in a new windowLink opens in a new window / UoW OA Research RepositoryLink opens in a new windowLink opens in a new windowLink opens in a new window]

5GNILU

5G technologies

Mobile telecommunication operators, infrastructure suppliers, car manufacturers, and local councils are all seeking to understand the benefit from the leap in bandwidth promised by 5G technologies, and are keen to use the very latest 5G evaluation technology.

mmWave

Our team acquired the UK’s most advanced diagnostic and testing platform for a key part of the 5G spectrum - mmWave. This technology promises to deliver a step change in the amount of data that can be wirelessly transmitted, opening up opportunities for a range of new services and products, including those associated with enabling connected and autonomous vehicles (CAVs). We are already working with a range of industrial partners on connectivity, and the understanding and optimisation of user/customer interaction with driverless technology. This new facility will further enhance our vision to be the UK's “go-to” CAV development platform providing unrivalled research and testing that will accelerate product introduction, infrastructure design and implementation. The technology developed will be transferable to other sectors beyond automotive.

We are currently integrating the new 5G mmWave technology platform into existing projects. Core to our methodology is the ability to evaluate early in the development cycle the potential use of 5G mmWave communications systems within project partners’ products. NIs leading platform provides us with both state-of-the-art hardware, capable of transmitting real information in the frequency bands of choice, as well as access to the control software, providing agility when being deployed in real scenarios. This allows us to be ahead of the CR&D curve and further enhances our standing as the preferred partner within the UK's CAV community.

Publications and Press Releases

[Journal]: "Millimeter-Wave Communication for a Last-Mile Autonomous Transport Vehicle" in IEEE Access [DOILink opens in a new windowLink opens in a new windowLink opens in a new window / UoW OA Research RepositoryLink opens in a new windowLink opens in a new windowLink opens in a new window]

[Press Release]: WMG has set an autonomous vehicle communications record using 5G [link]

[Press Release]: WMG has just acquired the UK’s most advanced diagnostic and testing platform for a key part of the 5G spectrum [link]

Connected and Autonomous Vehicles

Associated Intelligent Infrastructure (CAVIE)

Research in Connected and Autonomous Vehicles and associated Intelligent Infrastructure Environment is of vital importance. By enhancing intelligent symbiotic relationship between the two would help improve traffic related problems and road safety.

Challenge 

Traffic related problems such as congestion, travel delays, road safety and air pollution are age old problems which are mainly caused by rise in traffic demand with limited infrastructure. Our research focus on improving these problems by managing the flow of freight transportation which is considered as the backbone of the economy. To improve the handling and transportation of freight goods, previous studies have highlighted methods such as traffic management systems, routing systems and Intelligent Transportation Systems (ITS). But such measures still struggle with ever increasing flow of traffic.

CAV-F

This research tries to analyse the influence of Connected and Autonomous Freight Vehicles (CAV-F) in improving road traffic flow, congestion and travel times. By simulating road traffic at the Dartford-Thurrock Crossing Tunnel, Kent, UK, using PTV Vissim’s traffic simulation model, Connected and Autonomous Freight Vehicles (CAV-F) are driven alongside conventional light goods vehicles, to determine the feasibility of increasing the traffic throughput at the tunnel. The results show that with the use of CAV-F, the overall traffic flow was increased by ~33% (i.e. up to 7,000 vehs/hr) from the current flow of ~5,000 vehs/hr. With the reduction in the headway and standstill distance and increase in the scope of intelligent connectivity and traffic speed limit, the average congestion and travel time are reduced even at a higher traffic concentration. By analysing the results, it has thus been possible to highlight the benefits to traffic management and road utilisation by introducing CAV-F into our road network, in the long term.

The video shows the comparison of traffic flow at Dartford-Thurrock Crossing, Kent, UK. On the left, the real-world mandatory five closures scenario is simulated to see how these closures affect traffic flow at 5,000 vehicles/hr traffic volume. The simulation on the right shows the simulated scenario where all the conventional freight vehicles are replaced by CAV-F, at 7,000 vehicles/hr traffic volume. We could see that the traffic simulation using CAV-F performed better than the other one even with a 40% higher volume of traffic.

Dynamic Gap Generation – Mathematical Model

The study also proposes a novel mathematical model for generating dynamic vehicular gaps by applying dynamic speed changes to allow the platoon of Connected and Autonomous Dangerous Goods Vehicles (CA-DGV), to travel in isolation via the tunnel. This will ensure the safety of other road users in line with check-and-allow procedures whereby certain DGVs are escorted via a tunnel in isolation for the safety of other road users, as ADR regulations and tunnel category.

The model is verified against different road layouts leading to a tunnel in the UK, to ensure all possible driving scenarios are considered. The different road layouts are simulated using PTV Vissim traffic simulation software. The results showed that the model works effectively for all given layouts and desired gaps are generated between the platoon of CA-DGV and its preceding and following vehicles when the speed changes are applied at the reference locations, calculated using the mathematical model.

As an example, the video shows a typical road layout with two junctions joining the main road before the tunnel. Three platoons of CA-DGV travelling on three road merge as one with an appropriate distance between their preceding and following vehicles to travel via a 200 meters long tunnel in isolation.

ELWAG

Enhanced Assured Location Simulation Leveraging Wi-Fi and GNSS for Sensor Fusion (ELWAG)

The need for smart devices to have a highly accurate self-awareness of their own location and the location of other smart devices around them is becoming increasingly important.

The need for smart devices to have a highly accurate self-awareness of their own location and the location of other smart devices around is becoming increasingly important. Many devices rely on a singular location technology (typically GPS), which is one type of the wider ecosystem of Global Navigation Satellite Systems (GNSS). GNSS is degraded typically in urban areas where buildings and other cityscape features interfere with the signal. Wi-Fi signals though exist almost universally within these situations or dense urban areas and so there is a possibility of 'fusing' these signals with the GNSS signals to identify one’s location very accurately. However, for the concept of a hybrid Wi-Fi and GNSS system to be further developed, manufacturers need to be able to test such technology in a cost-effective, repeatable and safe environment.

In order to verify performance, a deterministic approach was adopted in modelling radio propagation effects such as path loss, diffraction, reflections, multipath and signal obscuration. Spirent’s Sim3D was used to develop a model of the University campus and Wi-Fi access points (APs) were placed in the 3D model as terrestrial beacons. The approach adopted was based on using the results of an extensive measurement campaign and an iterative RF model to evaluate the ability of the simulator to replicate or generate artificial Wi-Fi beacons.

This project delivered a Wi-Fi Access point (AP) + GNSS simulation that allowed for controlled, repeatable testing, e.g. including moving vehicles, obscuration, spoofing. The product developed was also tested and integrated with WMG’s 3xD simulator.

Publications

E. Adegoke, J. Zidan, E. Kampert, C. R. Ford, S. A. Birrell, and M. D. Higgins. Infrastructure Wi-Fi for connected autonomous vehicle positioning: A review of the state-of-the-art. Vehicular Communications, 20:100185, December 2019. ISSN 2214-2096. doi:10.1016/j.vehcom.2019.100185

E. I. Adegoke, J. Zidan, Elijah I. Adegoke, E. Kampert, Col R. Ford, S. Birrell, M. Higgins, P. Jennings “Evaluating Machine Learning & Antenna Placement for Enhanced GNSS Accuracy for CAVs”, IEEE Intelligent Vehicles (IV2019)

May 01, 2018 [Press Release]: News on our newly awarded Innovate UK Grant - ELWAG [link]

October 26, 2017 [Press Release]: News on our partnership with Spirent Communications Plc [link]

LIBConnect

Lithium-ion Cell-to-Cell Power Line Communication for Energy Storage Systems

Improvements to the fuel economy and rate of charge is instrumental in making Electric Vehicles a viable alternative to Internal Combustion Engine Vehicles. Electric Vehicles typically use Lithium‑ion (Li-ion) batteries to power their propulsion system due to their high energy and power density compared to other cell chemistries. However, significant research is required to understand the characteristics of Li-ion batteries in order to increase their performance, and consequently refuelling rates and range.

Established research demonstrates that cell instrumentation may be used to obtain enhanced cell characteristics. Such systems currently require a wired connection from the sensor embedded within the cell to the external data acquisition system. The addition of a wire harness within the battery pack and the increased risk of irreversible damage to the instrumented cell results in a regression of the performance enhancements expected from in-situ cell characterisation. An alternative method of data acquisition is therefore required for data acquisition from instrumented cells.

This project aims to produce a Power Line Communication system for use within an Electric Vehicle Li-ion battery pack. This system will take into consideration the significance of noise, power consumption, communication performance, and the environmental impact of prospective systems.

Design Method

This study will be exploring Power Line communication approaches for both cell‑to‑cell and cell‑to‑BMS systems. The benefits of in-situ Power Line communication system within an Electric Vehicle energy storage system includes the co-ordination of instrumented cells, allowing for unprecedented improvements in performance through cell reconfiguration and thermal protection.

Power Line Communication (PLC) allows simultaneous data and power transfer on the same wiring network. This project will seek to design a communication system that does not require significant changes to current battery pack designs. However, the effects of a system with coexisting Li-ion cells and a PLC system must be investigated.

Shown below are two constellation diagrams of a simulated quadrature amplitude modulated (QAM) signal with a carrier frequency of 1 MHz transmitted through two Li-ion cells in series. The Li-ion cells both attenuate and delay the signal transmitted through the cell. For a QAM signal with only two constellation points (2-QAM), which transmits two bits per symbol, the calculated bit error rate (BER) is theoretically 0. This is because the simulated attenuation, phase shift and noise applied to the signal is not significant enough for errors to occur. In contrast, when using 256-QAM the margin of error is greatly reduced, which increases the significance of the effects the Li-ion cells have on the signal so much that a BER above 98% is obtained.

The proposed in-situ Power Line communication system for Electric Vehicle battery packs within this project will be designed to mitigate the effects of the Li-ion cells on the signal characteristics of the system.

RACeD

Research for Advanced Concept Development

The RACeD, or Research for Advanced Concept Development, Doctorate Programme is a project funded by JLR with five Engineering Doctorate researchers investigating specific JLR research challenges. They will collectively advance JLR’s pathway to building up smart, connected and autonomous technology capability.

Evaluating the Self-Learning Car in the Simulated and Real-World

The future self-learning car must consider what features customers will want, what they will need and what features are worth automating. Perhaps most importantly we must consider how technology such as electric vehicles and self-driving cars will affect these user requirements. To this end, the direction of the research points towards the development of a future-proof framework for the development of self-learning and autonomous features within vehicles.

Predicting Pedestrian Movement using external perception sensors

In order to safely navigate a complex environment human drivers will observe their surroundings and make predictions on likely future states. Human drivers are able to create relatively complex predictions of other human's behaviour based on a complex set of information including features such as the local geography, body pose, face direction etc. Anecdotally many drivers can attest to being able to tell whether a pedestrian is planning to cross in front of them or not just by looking at them. Autonomous or semi-autonomous vehicles currently lack this complex prediction model, instead relying on simpler prediction models more prone to false positives. As higher-level autonomous vehicles become more common it is important for them to be able to smoothly interact with pedestrians. As such this project aims to explore whether a machine learning algorithm, utilising automotive sensors, can be trained to produce a more accurate prediction model for pedestrian movement when compared with current human-designed models.

A User-in-the-Loop Test Methodology for Wireless Network Services in Vehicles

The study will provide answers to the industry’s nonstandard use cases such as how much complexity is added when investigating connected services for a moving car as opposed to a static car and how important the user(s) activities for the in-car communication channel. The main objective of the project is to create a general model of the relationship between the wireless network parameters measured inside a moving car, possibly including its user interactions. This model will be useful as a prediction tool for assessing the perceived quality of in-car connected services.

Keeping the Driver Aware in a Semi-Autonomous Car

What happens if a self-driving car needs to hand back control to the driver, but the driver is distracted or unaware? This is a potentially challenging situation for the driver, vehicle and other road users alike. The research question is ‘How best can the car interact with the driver to keep them aware whilst the car is self-driving?’ This design challenge is the focus of this research and is split into three key areas: Safety, Interfaces and testing.

On-Board and Off-Board Data

In the automotive industry, there is a strong trend toward increased data output from the Electronic Control Units (ECU) in order to enable more complex features. This presents challenges that include: the bandwidth and topology constraints of the In-vehicle Network (IVN); the bandwidth and latency of wireless communication platforms (2G/3G/4G LTE) and the inability to effectively store, process and query the available information wherever it is needed. In order to make the most effective use of this data, this project aims to research techniques that can assist with data transfer and management between the On-Board and Off-Board systems.

A full-scale system integrating logistics with manufacturing operations is being implemented in the WMG International Manufacturing Centre. This system showcases Industry 4.0 methods and encompasses both new production systems and legacy equipment within a series of advanced manufacturing scenarios. The system is being used for both research and training with a range of industrial partners.