Deeraj Kumar Kanala

Creating intricate electronic control units (ECUs) tailored for advanced driver assistance systems (ADAS) and autonomous driving (AD) across a diverse range of automotive manufacturers within the industry's spectrum.

Requirement elicitation and management using DOORS. SysML/UML modelling using IBM Rhapsody. Establishing connections between components of the system(MBSE) Tested and validated ECUs using Vector tools (CANoe and CANape). Contributed to the Requirements analysis of ADAS control unit with stake holders. Created data-driven KPI dashboards using Power BI and Excel.

Topic: Multitasking learning for the dynamic safety of autonomous Vehicles, TU Kaiserslautern. Development of a unified neural network architecture to perform three different tasks. Performing ablative studies to test model behavior. Testing the model inference on different datasets (NYUDv2, KITTI, Cityscapes).

Training Neural Networks with real-world image data to detect traffic light from the scene. Used state-of-art algorithms and trained the neural networks using latest Deep learning Frameworks such as Tensorflow and PyTorch with GPU clusters. Developed existing Deep Learning tool-chain using Python. Evaluated and visualized the training and KPI metrics. Collaboratively worked together with USA and German teams.

Contributed to testing and validation of the product to meet customer requirements/standards. Assembly and testing of aircraft communication units (ECUs). Conducted comprehensive mechanical tests such as JIG testing, vibration, air, water, and humidity tests on ECUs.

Modelling and design of Various automobile components using SOLIDWORKS and Catia. Preparation of check sheets and DFMEA for the designed product. FE modeling for BIW , Plastic and Solid components, as well as part modeling and release. Worked with several other cross-functional teams on engine compartment design.

Interdisciplinary course fouces on Automotive, electrical, and Computer Engineering.