Lediga jobb Sensrad AB i Göteborg

About Sensrad
At Sensrad we build and develop one of the world’s most advanced radar sensors. Our sensor provides measurements of its surroundings containing thousands of four-dimensional detections in each frame, delivering a radar image of the environment with a uniquely high resolution. This, in combination with the fact that a radar works in any weather and light condition, has sparked great interest among customers. We supply a variety of verticals ranging from heavy vehicles and drones to static installations surveilling high-value infrastructure and intelligent traffic management. Today, we have customers in all these areas located all over the world.
Currently, we are looking for an industrial PhD student which in collaboration with our partner Örebro University conducts research which explores 4D imaging radar as a foundation for enabling autonomous operations in mining and construction vehicles to drive safely in all conditions.
The Project
The industrial PhD project described here is part of REEDEAM, a new graduate school led by Mälardalen University together with Örebro University and Luleå University of Technology. The program will train 15 PhD students as future technical leaders in areas such as automation, digitalization, metallurgy, and energy technology. In collaboration with leading companies including Epiroc, LKAB Minerals, Bosch Rexroth, Kanthal, RISE and Sensrad, REEDEAM aims to support Swedish heavy industry in reducing carbon emissions and driving the green transition.
The initiative builds on the REEDEAM Competence Development project and strengthens the link between academia and industry, creating long-term impact in carbon-free metallurgy, AI and automation, and circular industry solutions.
Your Role
As an Industrial PhD student, you will research, implement, and evaluate radar?centric odometry, localization, and SLAM algorithms, and integrate them into multimodal perception stacks for autonomous vehicles. The project targets autonomous driving for off?road and industrial settings (e.g., off-highway, mining and construction). You are expected to publish the research findings in top signal processing and robotics conferences and journals. Moreover, you will collaborate with Sensrad engineers and university representatives to help transition your research into products.
Required qualifications
To qualify for the position, you must have
A master’s degree in electrical engineering, computer engineering, applied physics, robotics or equivalent.
Be proficient in spoken and written English.
Solid skills in computer programming (Python / C++)
Experience with signal processing and perception algorithms (such as deep machine learning and Bayesian statistics methods.)

Meritorious qualifications
Experience with 4D imaging radars.
Knowledge about state estimation and SLAM methods.
Research publications or open-source contributions.

What we offer
We offer a unique opportunity to push the frontier of radar-centric autonomy on real applications and real deployments. You will work with a cutting-edge 4D radar sensor on diverse platforms and in challenging environments. Moreover, at Sensrad we have an international team and modern facilities. You may also travel to test sites and partner labs for collaboration, and we have a close collaboration with academia.
Supervision and Enrollment
This is an industrial PhD position co-supervised by Sensrad and Örebro Univeristy. Enrollment terms and coursework follow the university's doctoral program requirements. The expected duration is five years, where 80% of the time is dedicated for research contributions and 20% on activities at Sensrad. The position is in Gothenburg, but the student is expected to travel frequently to Örebro to collaborate with the research partner and to participate in courses at the university. This is a paid employment and is fully funded from the start.
How to apply
Please submit the following information:
CV
Degree transcripts (B.Sc. and M.Sc.)
Personal letter; introduce yourself, describe your previous relevant experience and your future goals and research focus.
Contact information for two references.
Submit your application no later than November 1st. 

Apply for this position via the career portal at career.sensrad.com. Visa mindre

Sensrad is a growing start-up and we're currently on the hunt for talented software developers. You will be part of our highly skilled software and perception team at our office located in Gothenburg, Sweden. The team is working on software development and perception algorithms for our state-of-the-art 4D imaging radar sensor. Our aim at Sensrad is to provide sensing capabilities that is at the cutting-edge of what is currently technical possible.

We believe that you possess a M.Sc. degree in computer science, software engineering or similar. Moreover, we think that you are used to working in an adaptive environment and are flexible towards new challenges. At Sensrad, we see our self as committed, self-driven and passionate about driving things forward.

Your work will involve.

- C/C++

- Python

- Embedded systems

- Docker

- Linux scripting

- Hardware-in-the-loop

- ROS2

- Bash, GIT, Gerrit, Jenkins

- OpenCL/CUDA

- ASPICE, Functional Safety (ISO 26262) and Cyber Security (ISO/SAE 21434)

You need to be familiar with several of these topics and be an expert at some.

Our customer and partners are located worldwide, and the position may require occasional travels, meetings and on-site tests.

If you're looking for a new challenge and want to make a real impact, this is the opportunity for you!

Closing date is 17th of November, but we work with ongoing selection. For communication and quality assurance reasons, we ask you to send in your application digitally via our system and not via e-mail. Visa mindre

Figure: One example of our 4D radar point cloud and the corresponding camera image showing four pedestrians.

Background

Given a 4D radar point cloud of an environment, it is typically not obvious which points belong to certain objects and which type the object is. Hence, there is a need to develop algorithms to obtain these object detection and object classification capabilities. For other sensors doing these tasks, such as camera and lidar, the advancements in the field of deep machine learning have been utilized to revolutionize the performance of such algorithms. However, for radar sensors, this leap in performance is yet to be taken. With this master’s thesis work, we aim to bring these revolutionizing deep neural networks to our radar sensor.

To be able to train deep neural networks which have groundbreaking potential it is generally required to have a big data set of annotated data points. However, to annotate data is both an expensive and a time-consuming process. A training strategy that utilizes pre-existing annotations, or decreases the overall need for annotations, is therefore of high interest.

Goal of the thesis
The aim of this master’s thesis is therefore to investigate deep neural network training strategies that can use both available public data sets of 4D imaging radar point clouds together with point clouds generated by our Sensrad radar sensor. Moreover, a stretched goal of the work is also to investigate neural network architectures that are able to train in a self-supervised fashion to decrease the need of annotated data. We believe that a transformer architecture could be one interesting candidate for such architecture.

This thesis shall consider:

- Select suitable public data sets to include in a training pipeline for object detection and/or object classification using deep neural networks.

- Perform a literature study to find the most promising state-of-the-art deep neural networks that can be trained on multiple 4D radar sensors (i.e. the ones used by the public data sets and our Sensrad 4D radar sensor).

- Explore how to train the neural networks in a self-supervised fashion.

- Evaluate the quality of the developed transfer learning methods on a selected inference task (e.g. object detection and/or object classification), and investigate if the selected method can run in real-time on our radar system.

Our radar sensor is one of the most advanced in the world and we think this thesis has the potential to make a significant contribution to both the research community and to our customers. We would therefore be happy if your achievements could lead to a publication and make it into our product. Visa mindre

Sensrad is a growing start-up and we're currently on the hunt for talented software developers. You will be part of our highly skilled software and perception team in our office located in Gothenburg, Sweden. The team is working on software development and perception algorithms for our state-of-the-art 4D imaging radar sensor. Our aim at Sensrad is to provide sensing capabilities that is at the cutting-edge of what is currently technical possible.

We believe that you have possess a M.Sc. degree in computer science, software engineering or similar. Moreover, we think that you are used to work in an adaptive environment and are flexible towards new challenges. At Sensrad, we see ourself as committed, self-driven and are passionate about driving things forward.

Your work will involve.

- C/C++

- Python

- Embedded systems

- Docker

- Linux scripting

- Hardware-in-the-loop

- ROS2

- Bash, GIT, Gerrit, Jenkins

- OpenCL/CUDA

- ASPICE, Functional Safety (ISO 26262) and Cyber Security (ISO/SAE 21434)

You need to be familiar with several of these topics and be an expert at some.

Our customer and partners are located worldwide, and the position may require occasional travels, meetings and on-site tests.

If you're looking for a new challenge and want to make a real impact, this is the opportunity for you!

Closing date is 7th of November, but we work with ongoing selection. For communication and quality assurance reasons, we ask you to send in your application digitally via our system and not via e-mail. Visa mindre

Figure: On example of our 4D radar point cloud and the corresponding camera image showing four pedestrians.

Background

Given a 4D radar point cloud of an environment, it is typically not obvious which points belong to certain objects and which type the object is. Hence, there is a need to develop algorithms to obtain these object detection and object classification capabilities. For other sensors doing these tasks, such as camera and lidar, the advancements in the field of deep machine learning have been utilized to revolutionize the performance of such algorithms. However, for radar sensors, this leap in performance is yet to be taken. With this master’s thesis work, we aim to bring these revolutionizing deep neural networks to our radar sensor.

To be able to train deep neural networks which have groundbreaking potential it is generally required to have a big data set of annotated data points. However, to annotate data is both an expensive and a time-consuming process. A training strategy that utilizes pre-existing annotations, or decreases the overall need for annotations, is therefore of high interest.

Goal of the thesis
The aim of this master’s thesis is therefore to investigate deep neural network training strategies that can use both available public data sets of 4D imaging radar point clouds together with point clouds generated by our Sensrad radar sensor. Moreover, a stretched goal of the work is also to investigate neural network architectures that are able to train in a self-supervised fashion to decrease the need of annotated data. We believe that a transformer architecture could be one interesting candidate for such architecture.

This thesis shall consider:

- Select suitable public data sets to include in a training pipeline for object detection and/or object classification using deep neural networks.

- Perform a literature study to find the most promising state-of-the-art deep neural networks that can be trained on multiple 4D radar sensors (i.e. the ones used by the public data sets and our Sensrad 4D radar sensor).

- Explore how to train the neural networks in a self-supervised fashion.

- Evaluate the quality of the developed transfer learning methods on a selected inference task (e.g. object detection and/or object classification), and investigate if the selected method can run in real-time on our radar system.

Our radar sensor is one of the most advanced in the world and we think this thesis has the potential to make a significant contribution to both the research community and to our customers. We would therefore be happy if your achievements could lead to a publication and make it into our product. Visa mindre