Darpan Antala

Experienced engineer with a solid background in reliability improvement and facility management as well as extensive expertise in utilizing resource planning tools such as SAP, MS Project and Primavera. Possesses exceptional
management and negotiation skills, including leading cross-functional teams and ensuring safety compliance. compliance with safety regulations. Proficient in vehicle engineering, data analysis, project management and creation of documentation to support project execution. Adept at optimizing maintenance activities to increase
efficiency and cost-effectiveness and in carrying out HiL and MiL simulations. Experienced in the development of new business models, strategy development and process digitalization in the automotive sector.

Generation and Identification of Edge-Case Scenarios in Automated Driving Systems

• Implementation of a Method for Automating the Identification of Edge Cases in Autonomous Driving Datasets Using Machine Learning and Anomaly Detection
• Combination of Extended Isolation Forest (EIF) and DBSCAN in a Hybrid Strategy to optimize anomaly detection and enhance performance.
• Conducting Multidimensional Anomaly Detection with EIF to address complex real-world driving scenarios.
• Proficiency in Advanced Machine Learning Algorithms for Anomaly Detection.
• Experienced in Handling and Processing Large Datasets for testing and validation of autonomous vehicles.

• Conducting deep learning-based semantic image segmentation on camera images with augmentation policies and object detection for publicly available datasets using TensorFlow, NumPy, and Pandas libraries.
• Implementation of inverse sensor models for creating occupancy grid maps from Lidar point clouds.
• Experience with neural networks and machine learning methods using tools like PyTorch, TensorFlow, Keras, and scikit-learn.
• Building a multi-class classification model using Keras's functional API and applying transfer learning concepts to enhance model performance.
• Performing object detection and localization on Bird’s Eye View datasets using advanced transfer learning concepts.
• Improving model interpretation by implementing gradient-weighted class activation maps to investigate the underlying interpretation of the CNN architecture.

Design of battery pack for electric forklift truck

• Designed the battery pack and battery management system (BMS).
• Selected the appropriate Li-ion cell technology with the required calculations.
• Implemented BMS and thermal management to recover and store energy during deceleration, maximizing cycle time.
• Selected an operating strategy to maximize cycle time and calculated overhead and operating costs to determine cost recovery

Effect of tool pin profile and dwell time on Friction Stir Spot Welding (FSSW) of similar AA5083-T6 sheets, Research paper, Designed a unique FSSW tool with a triangular shoulder, replacing the cylindrical shoulder., Reduced dwell time by 50% with lower peak temperature compared to the cylindrical shoulder., Presented at the International Conference on Advances in Engineering and Technology 2018 in Dubai by EIRAI.

• ACDC: Automated Connected and Driving Challenges RWTH Aachen University / edX 10/2022 – 03/2023
• Introduction to TensorFlow for Artificial Intelligence, Machine Learning, and Deep Learning DeepLearning.AI
• Machine Learning Course: Hands-on Python in Data Science (Udemy, Ongoing) Self-Driving Cars specialization. (Coursera, ongoing)