Doctoral Research Fellow - KI for Robotisation in Forestry

Key information

The scholarship position aims to learn robots to understand the forest well enough to be able to navigate and move through it in real time. This will be achieved by means of machine learning on Lidar point clouds and real-time understanding of the forest environment, and by exploiting this understanding of localization and navigation. A robot builds up a picture of their surroundings with a Lidar scan at a time. Each scan provides limited information, but overall they describe the forest in detail. Traditional methods of analysis and inference are designed to work with completed and quality-assured data set. On the other hand, a robot must be able to make estimations based on incomplete observations while moving it.

The PHD project aims to clog this gap by further developing machine learning models for segmentation and interpretation of forest clouds to real-time-based, incremental estimation systems. The work will support navigation, self-location and assessment of accessibility in forest terrain.

A little about the frames for the position

The position is a three-year doctoral position with workplace at Nibio's headquarters on Ås, 30 Km south of Oslo. The candidate will be admitted to a doctoral program at the Norwegian University of Life Sciences (NMBU) and will work closely with the departmental field robots and associated research activities.

In order to qualify for the position, the applicant may not have previously had a fellow position at NIBIO. The applicant may also have achieved a doctorate or previously held a fellow position financed by the Norwegian Research Council.

Applicants who do not document skills in Norwegian, Swedish or Danish at the minimum level A2, will be offered free Norwegian education through NIBIO. It is expected that the candidate achieves Norwegian knowledge at least this level during the employment period.

Position's main tasks

• Developing machine learning architectures that interprets 3D point show data incremental for real-time use, and which gradually builds up an understanding of the forest environment as new observations become available.
• Explore compact forest structure representations and investigate how these can support localization, detection of places and understanding of the surroundings.
• Customize models to the properties of real-time sensor data, including sparse and uneven sampled data, motion distortion and partial observations.
• Implement and evaluate the developed systems on built-in platforms and on Nibio's robot platforms.

Kvalifikasjonskrav

• Masters degree or equivalent five-year higher education in computer science, machine learning, robotics or a related subject area. The degree must be completed before accession in the position.
• Qualifications that meet the admission requirements of the relevant doctoral program at the Norwegian University and Life Sciences (NMBU).
• Experience with implementation, adaptation and further development of deep neural networks and other machine learning models.
• Good programming skills in Python and experience with modern framework for deep learning, such as Pytorch or TensorFlow.
• Good skills in English, both oral and written.
• Will and physical capacity to participate in field work under varying conditions in Norwegian forest areas.

It is an advantage if you have

• Deep learning for 3D data and sensor data, including spot clouds, voxel and savings representations, transform models, segmentation and registration. It is particularly desirable with experience from the development or adaptation of model architectures, beyond pure application of existing solutions.
• Real-time information and resource-limited systems, including quantization, pruning, model compression (distillation), cuda and built-in accelerators. Experience with C ++ or other systemary programming is an advantage.
• Robotics and state estimation, including sludge (simultaneous localization and mapping), sensor fusion, detection of places (Place Recognition) and Praise / Ros2. Experience with unmanned aircraft (UAV), mobile robots or machines in agriculture and forestry will be emphasized.
• Practical experience with development, integration and use of physical sensor systems, as well as planning and implementation of field trials.
• Knowledge of forestry, remote sensing or ecology.
• Knowledge in Norwegian or another Scandinavian language.

Personal characteristics

• Having the ability to work independently with complex and open research problems, from ideas development and method design to implementation, testing and validation in fields.
• Have strong analytical skills and ability to quickly familiarize themselves into new methods, tools and disciplines.
• Is motivated for and engaged in contributing to an ambitious and interdisciplinary research environment.
• Have good cooperation and communication skills, and the ability to interact effectively with colleagues and external partners with different academic backgrounds.

We offer

• Ability to work with an important social mission focusing on sustainability and forward-looking projects
• Challenging, exciting and varied tasks in interdisciplinary teams
• A flexible everyday with the opportunity to influence their own work tasks
• A strong professional environment with good colleagues
• A generous personal policy with welfare schemes and focus on the employees' different life phases

The position is placed in government regulations as a fellow, job code 1017, annual salary NOK 555.000 - 635.000. Remuneration is considered after education and experience.

Membership in the Government Pension Fund, which entails a good occupational pension scheme, occupational injury and group life insurance, and offers for mortgages to favorable interest rates.

More about NIBIO as an employer can be found here:Jobs in NIBIO

How to apply for the position

Application with CV is filled in electronically and sent via link on this page, " search position ".
Diplomas and testimonials are included in any interview, or are electronically placed.

NIBIO conducts background checks on current candidates.

Other information

We emphasize diversity in the workplace and encourage qualified candidates to apply - regardless of age, gender, functional ability, holes in CV, national or ethnic background.
As state business, we have the opportunity to provide a positive special treatment for applicants who have impairment, holes in the CV or immigrant background. Positive special treatment involves, among other things, that you will more easily come to interview.
Read more about thishere.

We note that information about the applicant can be made public, cf. Act on the right to access documents in public activities (Public Glegova), § 25. An applicant may request to be exempt from entry on public search list. If the request is not included, the applicant is notified of this.