# Event Details - **Event Name**: Technical Presentation - **Event Start and End Date**: Thu, 19 Feb, 2026 at 06:00 pm – Thu, 19 Feb, 2026 at 08:00 pm - **Event Description**: Technical Presentation MCAST Engineering Final Year Presentations 2025 Date: Thursday 19th February 2026 Time: 6.30pm Complimentary refreshments will be served at 6.00pm Location: The Embassy Hotel, Valletta Predicting Pedestrian Behavior as a Critical Safety Mechanism in Autonomous Vehicles – Abela Luke The study investigates pedestrian intent as a critical safety mechanism in autonomous vehicles by using Artificial Intelligence. The research focuses on applying deep learning frameworks in a multistage pipeline. The pipeline is composed of three algorithms, including YOLO as the main object tracker used for pedestrian detection, Deep SORT for identity tracking of each pedestrian and an LSTM model for predicting pedestrian intent. The system was trained and evaluated on the JAAD dataset, which was designed for studying joint attention analysis in autonomous driving. The experimental results showed strong performance in pedestrian crossing classification, especially in dense and high traffic environments. Across 100 videos, the model achieved an Intersection over Union score of 88.32%, an F1-score of 96.02% and an overall classification accuracy of 95% as shown by the confusion matrix. Luke Abela 3 While the results confirmed the pipeline was satisfactory, some limitations were observed during the experimental procedure, primarily due to the characteristics of the algorithms used and some constraints in the dataset utilised both of which affected the overall results of the system. Furthermore, future work involves incorporating additional features to fine-tune the algorithms, thus improving classification across a broader diverse set of scenarios and ultimately enhancing system flexibility. In conclusion, the study emphasizes the significance of predicting pedestrian intent as a crucial factor in ensuring autonomous vehicle safety. The findings contribute to the knowledge found in terms of the results obtained and thus conclude that the system obtained 95% against the JAAD dataset, ultimately providing a foundation for future advancement and real-world deployment of autonomous driving systems. Studying the Effects of Swirl-Induced Flow by Modifying the Intake Manifold of an Internal Combustion Engine – Cassar Aryton This project investigates how modifying the air intake of a small, four-cylinder engine in a city car can improve performance and efficiency. The study focuses on the use of twisted ducts, known as swirl generators, installed in the intake manifold to create controlled turbulence in the incoming air. The resulting swirl improves the mixing of fuel and air, leading to more stable and efficient combustion. Two designs were tested: a quarter-turn and a half-turn twist. The effects of these designs were analyzed using CFD simulations to study airflow patterns, turbulence, and swirl propagation. These results were then validated by experimental testing on the engine using a chassis dynamometer. Power and torque figures were measured across the engine’s speed range, offering both numerical data and practical performance insights into the impact of the modifications. Results showed that the half-turn design increased low-to mid-range torque by approximately 9%, making the engine more responsive in city driving conditions, while the quarter-turn design enhanced power at the upper end of the engine’s rev-range by up to 10 HP, demonstrating that controlled turbulence can improve engine performance without electronic remapping or costly components. These findings highlight how relatively simple, low-cost modifications can make combustion engines more efficient and versatile whilst also providing a foundation for future research into adaptive airflow systems that could adjust swirl dynamically to match driving conditions and performance needs. Parameter Optimisation of Injection Moulding using PFA Material to Improve Dimensional Stability – Cassar Kurt This presentation explores the optimisation of injection moulding parameters for Perfluoroalkoxy Alkane, commonly known as PFA, a high-performance fluoropolymer used in precision sealing applications. The project was conducted in collaboration with Trelleborg Sealing Solutions Malta, where dimensional instability during moulding was creating assembly challenges and material waste, thus incurring significant costs. Due to PFA’s semi-crystalline structure and high processing temperatures, even small variations in parameters can significantly affect shrinkage, warpage, and final part dimensions. The focus of this research was to identify the most influential processing parameters and determine an optimal combination capable of consistently achieving the required tolerances for inner diameter, outer diameter, and axial height. To achieve this, a structured Design of Experiments approach was implemented, including screening through the Taguchi method followed by a Full Factorial analysis. Statistical modelling, mould temperature distribution analysis, and process capability evaluation were also used to further validate results. The findings demonstrate how systematic parameter optimisation can significantly enhance dimensional stability, improve repeatability, and reduce waste through a quality improvement. Ultimately, this research highlights the importance of data driven process control when manufacturing high performance polymer components Design of a High Dynamic Position-Controlled PMSM Drive – Farrugia Zack This dissertation details the design and implementation of a high-dynamic position-controlled drive for a Permanent-Magnet Synchronous Machine (PMSM). The primary objective is to achieve fast, precise, and responsive motor performance through Rotor Flux Oriented Control (RFOC), a technology essential for high-precision industries such as robotics, CNC machining, and medical equipment. The research follows a sequential methodology: mathematical modelling of the PMSM, followed by the design of a cascaded control architecture consisting of nested current, speed, and position loops. The hardware implementation features custom-designed PCBs, which include a three phase inverter, gate drivers, and Hall-effect current sensing systems, all interfaced with an STM32F407 microcontroller. To ensure accurate feedback, a 12-bit magnetic encoder was utilised alongside a Quadrature Phase Locked Loop (PLL) to mitigate quantisation noise and improve speed estimation. Experimental results validated the system's hardware and data acquisition capabilities. The current controller achieved a maximum measured bandwidth of 350 Hz. However, challenges were identified in the full closed loop implementation, where oscillations and instability occurred due to residual rotor misalignment. Despite these challenges, the study provides a robust foundation for future improvements, such as the integration of Extended Kalman Filters and the exploration of sensor less control strategies. ADC front end for a Cyclone 10LP – Zammit Christopher Modern high-speed Data Acquisition Systems (DAQ) rely on Field Programmable Gate Arrays (FPGA) for real-time processing due to their parallel processing capabilities, speed and reconfigurability. However, since FPGAs operate entirely in the digital domain, a disconnect is created when processing real-world analogue signals. This limitation highlights the need for an intermediate system capable of accurately converting analogue signals into a usable digital format that the FPGA can effectively process. The development of a low-noise Analogue Front End (AFE), as proposed in this project, addresses this disconnect by enabling the accurate digitisation of the signal, thus allowing the FPGA to process it in the digital domain. The AFE is able to concurrently digitise two analogue signals at a sampling frequency of 105MSPS with a 12-bit resolution. It integrates directly with an existing FPGA development board in a piggyback configuration, completing the mixed signal interface between the analogue and digital domains. The AFE was designed with modularity in mind, allowing the signal conditioning stage performed by the AAF to be implemented on a separate PCB, which is connected using SMA connectors. To support reliable high-speed operation, specialised PCB design strategies were employed, including length matching and layer stack-up planning. Extensive testing was performed on the analogue side of the AFE, and critical components such as the AAFs and the ADC driver were validated using appropriate testing techniques, confirming their expected performance. However, further testing on the digital side of the AFE and complete characterisation of the ADC's performance remain as important areas for future work. This is a free event open to the general public. - **Event URL**: https://allevents.in/valletta/technical-presentation/200029693972445 - **Interested Audience**: - total_interested_count: 0 ## Ticket Details - **Ticket URL**: http://facebook.com/200029693972445?utm_source=AllEvents.in&utm_medium=event-discovery-platform&utm_campaign=valletta-events ## Event venue details - **city**: Valletta - **state**: MA - **country**: Malta - **location**: The Embassy Valletta Hotel - **lat**: 35.89891 - **long**: 14.51244 - **full address**: The Embassy Valletta Hotel, 173, Strait Street,Valletta, Malta ## Event Organizer details - **affiliate_id**: 2 - **organizer**: MGPEI (https://allevents.in/org/mgpei/16994884) ## Event gallery - **Alt text**: banner - **Image URL**: https://cdn-az.allevents.in/events7/banners/318de796db2a53fa915f0b5b949d5728e582ececc920c6338f9a09cf0ca9f979-rimg-w1200-h445-dcffffff-gmir?v=1771506345 - **Alt text**: thumbnail - **Image URL**: https://cdn-az.allevents.in/events7/banners/54e6c0aad128f74b5bfb1bda7472548838250723697ad980093935540b2cefbc-rimg-w300-h300-dcffffff-gmir?v=1771506345 ## FAQs - **Q**: When is the event happening? - **A:** Thu, 19 Feb, 2026 at 06:00 pm - **Q**: Where is the event happening? - **A:** The Embassy Valletta Hotel, 173, Strait Street,Valletta, Malta - **Q**: Who is organizing the event? - **A:** MGPEI - **Q**: Where can I find ticket details about Technical Presentation ? - **A:** You can find ticket details about this event on AllEvents. ## Structured Data (JSON-LD) ```json [ { "@context": "https://schema.org", "@type": "Event", "name": "Technical Presentation", "image": "https://cdn-az.allevents.in/events7/banners/318de796db2a53fa915f0b5b949d5728e582ececc920c6338f9a09cf0ca9f979-rimg-w1200-h445-dcffffff-gmir?v=1771506345", "startDate": "2026-02-19", "endDate": "2026-02-19", "url": "https://allevents.in/valletta/technical-presentation/200029693972445", "location": { "@type": "Place", "name": "The Embassy Valletta Hotel", "address": { "@type": "PostalAddress", "streetAddress": "173, Strait Street,Valletta, Malta", "addressLocality": "Valletta", "addressRegion": "MA", "addressCountry": "MT" }, "geo": { "@type": "GeoCoordinates", "latitude": "35.89891", "longitude": "14.51244" } }, "eventAttendanceMode": "https://schema.org/OfflineEventAttendanceMode", "description": "Technical Presentation\n\nMCAST Engineering Final Year Presentations 2025\n\nDate: Thursday 19th February 2026\n\nTime: 6.30pm Complimentary refreshments will be served at 6.00pm \n\nLocation: The Embassy Hotel, Valletta\n\n\nPredicting Pedestrian Behavior as a Critical Safety Mechanism in Auton", "organizer": [ { "@type": "Organization", "name": "MGPEI", "url": "https://allevents.in/org/mgpei/16994884", "description": "The MGPEI represents three of the leading British Engineering Institutions:\n \n- The Institution of Engineering and Technology (IET)\n- The Institution of Mechanical Engineers (IMechE)\n- The Institution of Civil Engineers (ICE)" } ] } ] ```