Industrial PhD - Learning-Based Methods for Localization and Map
Scania CV AB / Elektronikjobb / Södertälje
Observera att sista ansökningsdag har passerat.
Visa alla elektronikjobb i Södertälje,
Salem,
Ekerö,
Nykvarn,
Botkyrka eller i
hela Sverige Visa alla jobb hos Scania CV AB i Södertälje,
Nykvarn,
Huddinge,
Järfälla,
Sundbyberg eller i
hela Sverige Scania is now undergoing a transformation from being a supplier of trucks, buses and engines to a supplier of complete and sustainable transport solutions.
Autonomous vehicle development at Scania is advancing at a very high pace and self-driving trucks and buses on public roads will soon be commonplace. Autonomous Transport Solutions (ATS) Research at Scania is responsible for developing, testing and piloting future frontier autonomous driving concepts.
Background
Accurate localization and mapping is an essential ability of a robot to keep track of its location in an unknown environment. Existing methods to achieve this task are global positioning systems, inertial navigation systems, lidar-based algorithms and vision-based methods, among others. Moreover, high-definition (HD) maps contain a-priori information of the environment, such as lanes, traffic signs, barriers, etc. HD maps complement the sensory information beyond the robot perception range, thereby increasing navigation safety and robustness.
Learn to localize
Vision-based methods work by analysing recognizable features in the acquired camera streams. There are many established feature detection methods for active sensors, including both hand-crafted and recently, deep-learning based. Even though state estimation theory is a well-studied field, the introduction of new sensing modalities with autonomous driving, together with the unprecedented advances in deep-learning methods makes it a novel and exciting research field. The goal of this project is to investigate the interplay between the task-based learning approach and probabilistic state estimation by exposing the learning agent to raw sensor measurements while being aware of complementary sensing modalities.
Lifelong Mapping
A lifelong mapping system is key for the long-term deployment of mobile robots in changing environments. The goal of this project is to identify the changes in the semantic information contained in the HD map based on the robot's sensory observation. The data collected through the lidar and camera pipelines are matched and associated with the a-priori map information using machine learning techniques and statistical methods. The key idea of this project is to build a decision process to update the HD map based on the data association process.
Both projects provide the possibility to pursue a first-principles thinking approach, beginning with raw sensor measurements and designing a self-evolving system that can formulate interconnections between sensors and a-priori information in a data-driven fashion, while overcoming the shortcomings of the individual sensor modalities.
To achieve these objectives, two industrial PhD students will pursue multi-disciplinary research in the areas of statistical filtering, deep learning, reinforcement learning, graph theory, optimization and sensing technologies.
Your profile
To apply, you should have a master of engineering degree (or equivalent) in applied mathematics, computer science, machine learning, robotics, engineering physics, electrical engineering or in a related technical science or engineering subject. We welcome applicants that are currently completing the master thesis project.
The successful candidate has solid mathematical knowledge and proven programming skills. We look for candidates that are self-motivated, engaged, and that can communicate in a pedagogic way to expert and non-expert audiences. You exhibit strong analytical skills, are well-organized and able to autonomously plan and execute tasks. You are fluent in English both in writing and in speech.
Requirements
• Solid theoretical knowledge of system theory, filtering, state estimation, machine learning.
• Experience with deep learning frameworks and architectures
• Passion for coding and demonstrating theoretical results through running software
Extra meritorious if you have experience in one or more of the following:
• Developing within a Linux-based development environment
• Solid Python, C++, and system building skills
• Knowledge and experience within other programming languages and/or MATLAB/Simulink along with a keen interest in software development
• A proven interest in interdisciplinary research and record of initiative
• Experience with Natural Language Processing
• Experience with Reinforcement Learning
Contact information
For more information, you are welcome to contact Laura Dal Col, at +46 (0)8 553 851 20.
Application
Apply with CV, cover letter and copies of your education certificate and transcript. The final application date is March 25, 2022. Interviews will take place continuously during the application period. A background check might be conducted for this position.
IMPORTANT: APPLICATIONS WITHOUT TRANSCRIPT OF RECORDS WILL NOT BE CONSIDERED
We are looking forward to your application!
Scania is a world-leading provider of transport solutions. Together with our partners and customers we are driving the shift towards a sustainable transport system. In 2020, we delivered 66,900 trucks, 5,200 buses as well as 11,000 industrial and marine power systems to our customers. Net sales totalled to over SEK 125 billion, of which over 20 percent were services-related. Founded in 1891, Scania now operates in more than 100 countries and employs some 50,000 people. Research and development are mainly concentrated in Sweden. Production takes place in Europe and Latin America with regional product centres in Africa, Asia and Eurasia. Scania is part of TRATON GROUP. For more information visit:
www.scania.com.Varaktighet, arbetstid
Heltid/ Ej specificerat
Publiceringsdatum2022-03-17ErsättningEnligt överenskommelse
Så ansöker duSista dag att ansöka är 2022-03-25
Klicka på denna länk för att göra din ansökanFöretagScania CV AB
Omfattning Detta är ett heltidsjobb.
Arbetsgivare Scania CV AB (org.nr 556084-0976),
https://www.scania.com/world/#/ Arbetsplats Scania
Jobbnummer 6444948
Observera att sista ansökningsdag har passerat.