Development of a Vehicle-Agnostic Direct-Sensing Telematics System for Electric Vehicles

Electrifying road transport offers significant environmental and economic advantages. However, South Africa’s unique transport landscape and competing developmental priorities demand that any diversion of limited public funds be justified by demonstrable economic and social viability. Achieving this requires accurate modelling of an electrified road network, supported by data linking operational behaviour and energy demand across vehicles. Yet South Africa lacks locally generated empirical datasets on electromobility, and no existing telematics platform offers universal applicability across electric vehicles (EVs) without extensive setup time. To address this gap, this work developed and validated a retrofit, vehicle-agnostic telematics platform capable of directly and non-invasively sensing traction-battery and mobility parameters. The measurements are transmitted securely via LTE Cat 1 bis using a lightweight messaging protocol, MQTT, with backlog retransmission, and presented through an online dashboard. The system integrates HVDC voltage and bidirectional inductive current measurement, GNSS-based motion tracking, accelerometry, and temperature measurement on a compact printed-circuit assembly within an IP67-rated automotive enclosure. It safely measures DC voltages up to 880V and bidirectional currents of ±1.5 kA, with galvanic isolation, basic insulation, and integrated transient protection. Data aggregation, local storage, and remote transmission occur at 1 Hz, with additional backlog retransmissions, while remote interfaces enable secure processing, storage, and visualisation of fleet data. Laboratory, EV, and on-road testing confirmed full compliance with the defined functional and non-functional requirements. Voltage and current sensing achieved subpercent full-scale accuracy across the −5 ◦C to 55 ◦C temperature range, while GNSS maintained typical positional and velocity accuracies of 1.5m and 0.05ms−1, respectively. Temperature sensing, accelerometry, and battery-backed timekeeping met their specified performance criteria for meaningful inference, and the system remained fully operational with real-time coverage under vibration and non-ideal supply conditions during high-speed automotive testing. The validated platform provides a practical, non-invasive, and accurate means of rapidly generating synchronised electromobility datasets under South African operating conditions. These datasets provide the empirical foundation for credible modelling of an electrified transport network, enabling the state to evaluate, design, and prioritise interventions for South Africa’s transition to electric mobility.

Grantor dc:publisher

Stellenbosch : Stellenbosch University

Year dc:date.issued

2026

Author dc:creator

• Shaboodien, Sameer

Advisor dc:contributor.advisor

• Booysen, M. J.

Language dc:language.iso

en

Repository record dc:identifier.uri

https://scholar.sun.ac.za/handle/10019.1/135860

OAI identifier oai:identifier

oai:scholar.sun.ac.za:10019.1/135860