BIO

My PhD research is dedicated to advancing pedestrian collision biomechanics through integrating artificial intelligence (AI), machine learning (ML), and human body modelling. By combining these cutting-edge technologies, I aim to enhance real-world safety outcomes, specifically by improving the understanding and mitigation of pedestrian collisions. Central to my work is developing an innovative in-situ pedestrian collision forensics tool. This tool, designed to assist forensic investigators, uses Bayesian optimisation to enhance the accuracy and efficiency of accident reconstructions. By offering detailed insights into the dynamics of pedestrian collisions in real-time, it provides a valuable resource for legal investigations and contributes to the improvement of safety-critical systems in transportation.

In addition, my research employs Physics-Informed Neural Networks (PINNs) and advanced modelling techniques to simulate the complex interactions between the human body and external forces during a collision. These models allow for a more precise prediction of collision impacts, supporting the development of enhanced pedestrian protection systems. The accuracy of these simulations plays a crucial role in refining vehicle safety designs and shaping future safety standards.

Alongside the forensics tool, I have also developed an in-situ injury assessment tool aimed at emergency medical services (EMS). This tool provides real-time predictions of injury severity based on collision data, helping first responders make informed decisions during critical post-accident care. By bridging the gap between accident reconstruction and emergency response, this tool serves as a practical, life-saving application that enhances the immediate care of pedestrians involved in accidents.

My research stands at the intersection of biomechanics, AI, and safety systems, focusing on translating complex computational methods into real-world solutions. Through the integration of AI and machine learning, I am able to model human body dynamics with greater accuracy, contributing to a deeper understanding of how accidents affect the human body. These insights are critical in developing new safety systems that reduce the risk of serious injuries or fatalities for pedestrians.

Furthermore, my work extends to the broader field of transport safety, where I contribute to the development of Advanced Driver Assistance Systems (ADAS) and the refinement of safety regulations aimed at protecting vulnerable road users. By improving the accuracy of accident predictions and injury models, my research directly influences the creation of smarter, safer transport systems. Ultimately, I am committed to advancing safety-critical systems in the transport industry, helping to create a safer environment for pedestrians and all road users.

Prior to my PhD, I spent five years as a Module Lead at Continental, Bangalore, where I led airbag ECU calibration projects using CAE and crash data analysis. I managed the calibration of pre-crash and airbag deployment algorithms across various vehicle models, optimising safety systems using MATLAB/Simulink. My work included developing AI-powered algorithms to enhance field event mapping and data quality, analysing large datasets to identify safety trends, and evaluating vehicle safety systems through simulations. I also coordinated with OEMs, suppliers, and research institutes, and streamlined testing processes by automating tasks with VBA and Python, significantly improving workflow efficiency.

Earlier, as an Engineer at QuEST Global, I designed and implemented control systems for industrial automation, boosting operational efficiency and automating engineering processes. I also gained foundational experience in programming at Cognizant, and developed R&D expertise in inverter manufacturing during my time at ARVI Systems and Controls. My combined experience in industry and academia has built a strong foundation in AI-driven safety systems, computational modelling, and vehicle safety technologies, with a focus on improving road safety outcomes.