Einride’s Post

Excited to share my Master's Thesis in collaboration with Einride! It has been an incredible journey exploring how to enhance safety for decision-making models in autonomous vehicles. Huge thanks to Georgios Kryparos for supervising me! My thesis, "Anomaly Detection in Autonomous Driving Systems for Identifying Adversarial Attacks, " will be published soon and marks the end of my Machine Learning studies at KTH Royal Institute of Technology.

When some of the world’s smartest scientists and some of the world’s brightest engineers pull together, there’s a good chance they’ll probably create some groundbreaking new technologies. To unleash this immense potential, Mercedes-Benz Research & Development North America, Inc. – our R&D hub in Silicon Valley – just announced new research collaborations with Stanford University and UC San Diego. Stanford has been recognised as one of the cradles of #AI innovation. Students at the Stanford University School of Engineering will conduct hands-on research in the field of AI and its applications in the automotive industry. The goal: help advancing the AI-driven passenger experience. UCSD is renowned for its innovative 3D imaging research. The Autonomous Vehicle Laboratory, part of the Contextual Robotics Institute at UC San Diego Jacobs School of Engineering, focuses on image perception and data fusion for automated driving. Innovative 3D scene reconstruction from camera images could ultimately contribute to enhancing the safety of automated driving features. Mercedes-Benz is providing funding and support from its R&D team for each collaboration. And we’ll put the students’ academic research to the test in real-world applications. We’re thrilled to work with the next generation of innovators in what we consider some of the most exciting and most promising fields one can launch a career in.#MercedesBenzGroup #CompanyCulture #Excellence #Innovation #CustomerFirst #Sustainability #Diversity #Integrity

🚗 Exciting Summer at the AS³ - Automotive Software Summer School! 🚗 This summer, I had the incredible opportunity to participate in the AS³ Automotive Software Summer School, and it was nothing short of amazing! Over the course of the program, I expanded my knowledge in key areas that are shaping the future of the automotive industry: 🔐 Automotive Security: Learned about securing automotive systems with one-time programs and explored enhanced TARA for connected and autonomous vehicles. 🚗 Assisted and Automated Driving: Gained insights into perception sensors, advanced driving functions, and objectification of vehicle dynamics. 🧠 Applied Machine Learning & Deep Learning: From advanced deep learning to machine learning-assisted programming, I developed hands-on experience with AI applications in vehicle technology. 🔋 Electric Propulsion & Power Control: Explored the electric propulsion system and advanced electric power control to better understand the move towards greener mobility. 💻 Simulation and Testing: Learned how simulation plays a critical role in ADAS/AD validation and verification, as well as automation in vehicle testing with advanced sensor technology. 🚀 Hackathon & Cybersecurity: Wrapped up the program with a stimulating hackathon and a deep dive into cyber risk management in the automotive sector. This experience has been invaluable in equipping me with cutting-edge skills for the future of automotive technology, and I’m excited to apply these insights in my professional journey! #Automotive #AI #MachineLearning #DeepLearning #Cybersecurity #ElectricVehicles #Simulation #Innovation #SoftwareEngineering #AutomotiveSecurity #Hackathon

Advancing self-driving cars using a remote driving station! 🚗 Meet Integrated Engineering Students Parker Gu, Tony Jiang, Rowan Zawadski, Mankanwar Singh and Harrison Narvey. They created a remote driving system for people to connect to and navigate self-driving vehicles, allowing them to avoid hazards beyond the algorithm. "While artificial intelligence excels at recognizing patterns and predicting events based on observed data, the situations encountered on our roadways are often unpredictable. Human intuition, judgment and the ability to understand social cues are still necessary to navigate complex scenarios." Learn more about their #UBCEngineering capstone project! ⤵️

I am excited to announce that I have embarked on an innovative project at the University of Florida under the guidance of Dr. Sandip Ray from the Department of Electrical & Computer Engineering at the Herbert Wertheim College of Engineering. Collaborating with my peer, Akhil Sai K, we are addressing a critical challenge in connected and autonomous vehicles (CAVs). Our project focuses on modeling both benign and malicious scenarios for CAVs during the merging process on an on-ramp. The core of our research aims to develop robust algorithms and employ machine learning techniques to tackle the issue of malicious attacks—specifically, the dissemination of false information among vehicles. By leveraging advanced sensor technology, our goal is to train these vehicles in a simulation environment to efficiently verify critical data such as speed, position, and the proximity of nearby vehicles. This will enable them to make safe and accurate decisions about whether to speed up or slow down in order to merge efficiently. This opportunity to combine machine learning with sensor technology to enhance the safety and efficiency of autonomous driving is not just a step toward academic advancement but also a stride toward future mobility solutions.

Our FY23/24 cohort of early career Test & Evaluation professionals are approaching the end of their training within our Nova Systems Test & Evaluation Centre of Excellence.   Teaching young professionals to become Test & Evaluation specialists is a fascinating process to be a part of.   Their ability to master digital skill sets such as model based systems engineering, physics-based modelling and simulation, and experimental design and analysis packages is amazing to watch.   Their mix of digital and classical T&E skillsets make them well suited for deployment onto complex T&E projects requiring innovative approaches.   Nova Systems Australia and New Zealand is now taking enrolments for the upcoming modules of our Advanced T&E Practitioner Training course which focuses on uncrewed and autonomous vehicle test methods:   · Test & Evaluation of Underwater Autonomous Vehicles (AUVs) (29 Apr - 3 May 24) · Scientific Methods for System of Systems T&E (Design of Experiments) (6 May -10 May 24)   These modules can be taken separately. #testandevaluation #WeAreNova #defence

Last week, I had the opportunity to present and discuss the core contributions of my research on model-based predictive control of autonomous marine vehicles (AMVs) at Industrial Systems and Control Ltd. The discussions were invaluable and will significantly influence my overall PhD research, aiming to better address motion control challenges under realistic operating conditions. As seen in the picture, the panel included renowned scholars in industrial control systems. From the right, going counterclockwise: Dr. Gerrit van der Molen, Prof. Ikuo Yamamoto, my humble self 🙂 and Prof. Mike Grimble.

Hello everyone... I am thrilled to present my latest Computer Vision project developed during my internship at AIMER Society - Artificial Intelligence Medical and Engineering Researchers Society under the guidance of Sai Satish sir. The project focuses on creating a traffic sign board detection system using the YOLO (You Only Look Once) algorithm. This system is designed to detect and classify various traffic signs in real-time, offering a solution that can be pivotal for applications like autonomous driving and smart traffic management. #APSCHE #MachineLearning #DeepLearning #YOLO #ObjectDetection #ComputerVision #ArtificialIntelligence #TrafficManagement #TechInnovation

TESTING: Remote vehicle testing from Indiana to Michigan From more than 320km away in Indiana, Purdue University researchers were able to test their AV technology using the University of Michigan (U-M) real-world test track in Ann Arbor, Detroit, using real and virtual cars, and pedestrians. Yiheng Feng, an assistant professor of civil engineering at Purdue, said, “The original path-planning algorithm was implemented and tested in our own vehicle here at Purdue. Through the Mcity 2.0 test, we were able to identify problems related to safety and comfort and refine our original algorithm, all without leaving campus.” Susan Margulies, National Science Foundation (NSF) assistant director for engineering, added, “NSF’s investment in computing and communication infrastructure for Mcity 2.0, together with the creativity of the Mcity 2.0 team, is upgrading what was previously a well-equipped test track to provide a unique, world-class research resource that is accessible for all campuses.” Read more here: https://lnkd.in/eJWEg3Gk #AutonomousVehicles #AV #SelfDriving #ADAS #AutonomousDriving #Automotive #AutomotiveIndustry #Transportation #Technology #FutureMobility #Sustainability #AVTExpoStuttgart #AVTExpoCA

I am happy to announce that I have completed the sensor fusion nano degree course from #udacity! 🎉 This course taught me the fundamentals of sensor fusion and perception for #selfdrivingcars. I learned how to use lidar, radar, camera, and Kalman filters to detect and track vehicles and pedestrians over time. I also worked on multiple capstone projects that involved building a complete #sensorfusion pipeline for autonomous vehicles. 🚗 My review of this course is that it is both challenging and rewarding. The curriculum covers a lot of theoretical and mathematical concepts, while also providing ample opportunities for practical, hands-on learning through industry-related projects. The instructors are real engineers who have developed sensor fusion algorithms for Mercedes-Benz cars, and their knowledge and expertise are evident throughout the course. One of the challenges that I faced during this course was that it required a lot more than just basic knowledge of C . C was the primary programming language used for most of the projects and exercises, and as a result, I had to brush up on my C skills and learn some advanced concepts. These included object-oriented programming, memory management, templates, and smart pointers. It was quite an intensive experience, but I was able to learn a lot and improve my skills in the process. 💻 Some of the essential skills and topics that I learned in this course are as follows: #lidar obstacle detection: I learned how to filter, segment, cluster, and classify lidar point clouds to identify obstacles on the road. 🚧 #camera: I learned how to calibrate, undistort, and transform camera images to extract useful information for perception tasks. 📷 #radar: I learned how to process radar signals, simulate an FMCW radar system, and apply various signal-processing techniques to detect a moving target. 📡 #kalmanfilters: I learned how to use linear and extended Kalman filters to fuse data from multiple sensors and use the constant velocity - constant turn rate model to estimate the state of moving targets. 🔮 If you're interested in learning more about my projects, please check out my 📁GitHub repository. https://lnkd.in/gTdkS2yM Additionally, I have also posted some 📝Medium articles related to these projects. https://lnkd.in/gYs_ycuu #udacity #sensorfusionnanodegree #autonomousvehicles