Pranjal Mujmer

I am a dedicated engineer specializing in Mechatronics and Cyber-Physical Systems, driven by a passion for advancing technology in automotive and robotics industries. With a Master of Engineering degree and a focus on LiDAR perception, my research delved into enhancing autonomous vehicle capabilities in adverse weather conditions.

My master thesis, "Analysis of Automotive LiDAR Raw Data and Perception Performance in Challenging Weather Conditions," explored the variability of LiDAR point cloud data under artificial rain and fog scenarios. This pioneering research, presented at AZ Live and published in a conference paper, contributed to significant strides in autonomous driving technology.

My expertise extends to developing synchronized simulations and immersive virtual reality experiences using Unreal Engine 4 and HTC Vive. I have integrated MongoDB, React framework, and PyQt 5 to automate tasks, showcasing proficiency in software development and innovation.

With a collaborative mindset and a commitment to lifelong learning, I thrive in environments where creativity meets technical excellence. I am eager to connect with industry professionals and explore opportunities to drive innovation and shape the future of technology.

m a Mechatronics and Cyber-Physical Systems engineer with a strong focus on sensor fusion, navigation, and 3D perception for autonomous and robotic systems. I specialize in ROS (1 & 2), C++, and Python, and have hands-on experience across the full development lifecycle—from sensor integration and data pipelines to testing, deployment, and real-time visualization.

Currently, I’m working as an ADAS Intern at AVL Software and Functions GmbH, where I’ve developed a high-performance C++/ROS 2 data parsing library and built GNSS/IMU diagnostic tools using real-time ROS topics and visualizations in RViz. As part of the EU-funded “Roadview” project, I’ve contributed to sensor anonymization with TensorFlow, calibration drift detection, and model versioning for sensor fusion pipelines.

For my Master’s thesis, I conducted over 50 controlled LiDAR experiments in adverse weather conditions, analyzing 20,000+ point clouds using PCL and validating object detection performance. I also implemented calibration-drift compensation methods to enhance perception reliability for autonomous vehicles.

In previous roles as a Working Student, I developed LiDAR–GNSS calibration pipelines using Ceres Solver, SLAM-based mapping, and multi-sensor calibration systems using Google Test and custom benchmarks. These tools helped improve localization accuracy and perception robustness under degraded conditions.

I’ve also led several robotics and simulation projects involving path planning algorithms, CNN-based segmentation, and point cloud clustering, and I’m experienced with both software development and mechanical prototyping (e.g., CAD, FEA, and assembly for SAE competitions).

I’m fluent in English and German, passionate about intelligent systems, and always looking to bridge hardware and software for real-world autonomous applications. My work has been published in peer-reviewed conferences, including on topics such as autonomous public transport and agricultural drone applications.

One of my most relevant technical trainings involved developing SLAM-driven LiDAR mapping and implementing LiDAR–GNSS calibration using Ceres Solver for bundle adjustment. This work focused on minimizing pose residuals and motion distortion in the ENU (East-North-Up) coordinate system, enabling high-precision localization and mapping in autonomous systems. It deepened my understanding of sensor fusion, optimization algorithms, and real-world coordinate transformations critical for reliable navigation.