Predictive Maintenance for Large Vehicles Fleets using Edge Computing Techniques
Cardiff University School of Computer Science and Informatics
Academic Contacts: Dr C Perera and Prof O F Rana
Project ID: CUK2E35
Annual Stipend: £14,628
Application Deadline: 21 February 2020
This PhD project focuses on developing a novel distributed predictive analytics technique that can efficiently be used in edge computing scenarios. Our partner iPoint is a company that develop sensor data based intelligent services (e.g., predictive maintenance), especially for large vehicle fleets such as trucks and busses. They provide a range of insights and recommendation to efficiently manage vehicles (e.g. evaluate driver behaviours and measure the quality of driving). Currently, all the sensor data collected by each vehicle and related accessories (e.g. sensor data generated by onboard sensors connected to the CAN bus) are directly sent to the cloud. All the required processing happens within the cloud, and relevant commands are sent back to each vehicle. This approach is inefficient from many aspects and could also impact the quality of service and customer satisfaction in certain scenarios.
• Sending data to the cloud is costly (i.e., mobile data plans could cost significantly when it is required to send large volumes of data to the cloud).
• Sensing data to the cloud and received the commands from the cloud (i.e.,g decision-making loop) has high latency (i.e., It takes time to send data to the cloud and receive commands back). Given that we are dealing with moving vehicles, even small latency could lead to negative incidents.
• Mobile network connections are unreliable; therefore, any service depends on mobile network connectivity could be unreliable.
• Depending on external connectivity also open up higher security risks (e.g., more vulnerable to cyber-attacks)
• Typically, network communication consumers more energy than data processing/analysis (locally)
All of the above problems can be fully or partially mitigated by developing and adopting edge computing methods.
Research Objective: In order to address the challenges face by iPoint, we aim to develop a novel data processing architecture that is capable of moving analytics across different nodes (within the architecture). This means that iPoint will no longer be required to send all the sensor data to the cloud all the time. The onboard computer will conduct most of the data analytics local and will only send the summarised/aggregated data to the cloud. However, our proposed algorithms will consider context information when deciding where the data analysis should happen.
Develop novel algorithms (ranging from predictive militances to driving behaviour analysis):
• Design and development of distributed algorithms that can dynamically orchestrate IoT resources on edge to satisfy a given sensing requirement without continuous connectivity to the cloud.
• These algorithms will also determine how to distribute data analytics workloads (among heterogeneous edge nodes) in an optimal way to satisfy given requirements and real-world constraints.
Develop novel data processing architecture:
• Design and develop self-organizing and reconfigurable IoT infrastructure that integrates resources from multiple layers (sensing, edge/fog, cloud).
START DATE: 1 APRIL 2020
ONLINE APPLICATIONS by 21 FEBRUARY 2020
In the funding field of your application, indicate applying for “KESS2 PhD Scholarship in Computer Science & Informatics”, and specify the project title and supervisors of this project in the text boxes provided.
UK tuition fees, stipend (£14,483 p.a. in first year – subject to confirmation), plus travel/conferences, support, consumables/equipment.
ELIGIBILITY – applicants must:
• have a home or work address in East Wales region (local authority areas Cardiff, Flintshire, Monmouthshire, Newport, Powys, Vale of Glamorgan and Wrexham) at application and enrolment;
• have the right to live and work in the UK for the duration of the scholarship, and the right to take up paid work in the East Wales region on completion of the scholarship;
• be classified as a ‘home’ or ‘EU’ student;
• satisfy the admissions criteria.
ACADEMIC CRITERIA: 2:1 Honours undergraduate or a master’s degree, in computing or related subject.
To be eligible, the successful candidate will need to be resident in the convergence area of Wales (West Wales and the Valleys) on registration, and must have the right to work in the region on qualification.
Knowledge Economy Skills Scholarships (KESS 2) is a pan-Wales higher level skills initiative led by Bangor University on behalf of the HE sector in Wales. It is part funded by the Welsh Government’s European Social Fund (ESF) convergence programme for West Wales and the Valleys.