Manthan Patel

- Developing functional safety software for object detection and tracking in Human-Robot Collaboration (HRC) applications using LiDAR technology. - LiDAR evaluation - Site security perception software development - Accelerated Computing

Topic: Exploring Deep learning approach for 3D object detection and 3D object tracking in industrial layout for functional safety application - Detecting and tracking humans in indoor industrial factory environment - Synthetic data generation using NVIDIA Omniverse Replicator and Isaac Sim - Deploymet of model on NVIDIA Jetson Orin - Study of Functional Safety standards using AI (ISO/IEC TR 5469) for the development and evaluation of the pipeline

Worked in Data Driven testing and Validation department for development of backend services of ComBox services used for data management and analysis of development vehicles. Mainly dealing with measurement files(mf4). Developed service for data management(filtering and labelling) to INFLUXDB. POC: Developed a service for generating automated test report containing measurement data from recorded files from the database. NLP: Worked on automatic code generator from text input of requirements.

Study on impact of augmentation on Neural Architecture Search Robustness

- Model development and Hyperparameter optimization of engine models consisting of airflow model, ignition advance optimization, emission optimization, performance optimization, idle calibration, in-use performance ratio(IUPR) and on-board diagnostics; based on testing data from test bench and vehicle. - Developed virtual test cycles for IUPR and RDE tests and validations. - POC project: Developed a machine learning model to predict catalyst temperature for catalyst temperature management.

Summer school on Automotive Software Engineering