ResearchSpace

Communities in ResearchSpace

Welcome to ResearchSpace, the institutional repository of the CSIR. ResearchSpace is an open access electronic archive collecting, preserving and distributing scholarly digital materials created by the CSIR.

Recent Submissions

Item

AI-BEI Portfolio

(2026-08) Business Excellence and Integration, CSIR

Artificial Intelligence's (AI) value is already being realised in the following ways: Productivity augmentation & reporting, Knowledge & information services, R&D, ethics & IP management. Its common value proposition: Speed + quality support + traceability + more time for professional judgement.

Item

Insights from ‘Unlocking COVID-19 current realities, future opportunities: Artificial intelligence in the time of COVID-19’

(2021-03) Mathe, Ntombizodwa R

The University of Cape Town, in partnership with Standard Bank, hosted a webinar entitled ‘Unlocking COVID-19 current realities, future opportunities: Artificial intelligence in the time of COVID-19’ on 19 August 2020. The webinar was facilitated by Professor Tommie Meyer from the University of Cape Town’s Centre of Artificial Intelligence Research at the Department of Computer Science. The two speakers for the event were Professor Tshilidzi Marwala, Vice Chancellor and Principal of the University of Johannesburg and a pioneer in the field of artificial intelligence (AI) in South Africa and the Deputy Chair of the Presidential Commission on the Fourth Industrial Revolution (4IR). The second speaker was Mr Nanda Padayachee, Head AI, Automation and APIs at the Standard Bank Group. He has extensive experience in digitisation of capabilities and established the Inaugural Data Science capabilities within the Standard Bank Group.

Item

Terrain-aware head gesture recognition for turret control using helmet-mounted IMUs and vehicle vibration fusion

(2026-09) Rooibaard, Lonwabo; Modungwa, Dithoto M; Sibiya, M; Pandelani, T

Head gesture-based control using inertial measurement units (IMUs) provides an intuitive alternative to conventional human–machine interfaces for mobile and vehicle-mounted systems. However, gesture recognition reliability degrades significantly under terrain-induced vibration and mechanically dynamic operating conditions. This study investigates terrain-aware head gesture recognition through the integration of helmet-mounted IMU measurements and vehicle vibration sensing to improve discrimination between intentional gestures and non-intentional motion artefacts. Vehicle vibration data were collected from a patrol vehicle traversing the Ndumo Border Patrol route, characterised by variable terrain roughness and dynamic excitation profiles. Triaxial seat-rack acceleration data were acquired at 10 kHz, anti-alias filtered and down sampled to 100 Hz before being integrated with IMU-derived head motion measurements using both vibration-aware data augmentation and early sensor fusion strategies. FFT-based spectral processing and classification using a lightweight fully connected neural network (FCNN) were implemented using the Edge Impulse framework. Experimental evaluation was performed using temporally independent training and testing segments to reduce overlap leakage and ensure realistic generalisation assessment. Results demonstrate that terrain-informed sensing substantially improves operational robustness under mobile conditions. The early-fusion approach achieved 98.45% independent-test accuracy under float32 inference and maintained 90.02% accuracy after int8 quantization, corresponding to an accuracy reduction of 8.43 percentage points. In comparison, the Clean IMU and vibration-augmented models exhibited reductions of 20.13 and 27.02 percentage points, respectively, demonstrating greater sensitivity to quantization. The evaluated models also exhibited low inference latency and compact memory requirements, supporting their suitability for real-time edge implementation. These findings demonstrate that treating terrain vibration as contextual information, rather than solely as environmental noise, can improve the quantization robustness and deployment characteristics of IMU-based gesture-recognition systems intended for mechanically dynamic platforms.

Item

A computational model for analysis of the dynamic response of a quarter-vehicle suspension system

(2026-09) Sikhakhane, Phakamani L; Joni, Emmanuel L; Mokgotho, Lebone Y; Modungwa, Dithoto M

Off road vehicle suspension performance directly affects ride behaviour, structural loading, and vehicle control under irregular terrain excitation. This paper presents a cross-platform computational framework for analyzing the vertical dynamic response of a quarter-vehicle suspension system using three modeling approaches: a reduced-order MATLAB model and multibody models in Project Chrono and MSC ADAMS. A common suspension definition and terrain input were used to compare sprung mass displacement, unsprung mass displacement, suspension travel, and sprung mass acceleration. MATLAB and Project Chrono show the closest agreement. while MSC ADAMS predicts a smoother, more damped response. The framework provides a practical basis for comparing modelling fidelity in quarter vehicle suspension analysis under off-road excitation.

Item

In-situ reactive synthesis and characterization of a high entropy alloy coating by laser metal deposition

(2022-03) Dada, M; Popoola, P; Mathe, Ntombizodwa R; Pityana, Sisa L; Adeosun, S

In this study, we investigate the influence of in situ reactive synthesis of Ti6Al4V– AlCoCrFeNiCu high entropy alloys by laser metal deposition on the microstructural and mechanical properties of the as-built alloy as opposed to the traditional method of mixing powders via a ball mill prone to contamination and segregation. We explore the capability of a new alloy design by combining two base alloys via in situ reactive alloying, delivering the Ti6Al4V and AlCoCrFeNiCu high entropy alloy powders from multiple powder feeders and regulating their feed rate ratios. The nano mechanical, tribological and microstructural morphologies of the alloys were characterized using a nanoindentation tester, a tribometer, XRD and SEM, respectively. The results showed that satelliting the high entropy alloys powder and the Ti6Al4V powder fraction using double powder feedstock had a homogeneous distribution with dendritic structures. Optimization was achieved at a laser power of 1600 W, a scan speed of 12 mm/s and a powder flow rate of 2 g/min. The surface roughness (Ra) for Ti–6Al–4V, AlCoCrCuFeNi and (Ti–6Al–4V)-(AlCoCrCuFeNi) alloy was 0.5 μm, 0.63 μm and 0.80 μm, respectively. The high wear resistance of the novel Ti6Al4V– AlCoCrFeNiCu alloy was influenced by the hardness of the alloy which was higher than the Ti6Al4V alloy and the AlCoCrFeNiCu alloy. This study successfully defines the capabilities of in situ fabrication of high entropy alloys and presents novel techniques for multiple powder preparation of high entropy alloys using laser additive manufacturing, to permit the next generation of compositionally graded materials for aerospace components.