+++ At DLR, we are always looking for bright minds – even during the current exceptional circumstances. Our current vacancies can be found here. +++

My knowledge is shaping the future.

Bookmark
Scientific activities / projects, PhD position

Development of algorithms for cooperative exploration of propagation processes with mobile robotic swarms

Starting date

immediately

Duration of contract

3 years

Remuneration

up to German TVöD 13

Type of employment

Full-time

Detection and mapping of dispersion processes, such as pollutants or gaseous substances in the air, is a very challenging but important problem in a variety of applications, such as monitoring and inspection of industrial facilities or environment protection or pollution monitoring. Typically, stationary sensor networks are used for such tasks. However, stationary sensor networks offer little flexibility in capturing highly dynamic nature of the gases being monitored.
 
The research group Swarm Exploration at the Institute of Communications and Navigation  investigates use of multiple cooperating robotic platforms - or swarms - to explore such processes. Specifically, we propose to use so-called in-situ as well as remote gas sensors to capture spatial gas concentration with using mobile robots.

The goal of this research is to

  • develop efficient exploration algorithms based on the measurements of all agents in the swarm, appropriate communication and navigation tools, as well as sensing technology, and
  • experimentally validate the developed algorithms.

The exploration algorithms should be based on information-theoretic considerations and use estimation techniques that are also widely used in communications engineering. The multi-agent system consists of flying and ground-based robots. Multiple robots are needed to handle the high spatial dynamics of the gas dispersion process. To support robot deployment and decision making, gas dispersion is represented with a physical model of gas concentration dynamics. This model also permits probabilistic formulation, so that the probabilistic distributions of the model parameters – be it locations of the gas sources or the spatial gas concentration map - can be inferred numerically from the measured values. In this context, the probabilistic approach to model representation plays an important role: it allows to compensate for unavoidable model mismatches and serves as a basis for the intelligent behavior of the multi-agent system.

The developed algorithms should exploit a decentralized architecture and generate optimal measurement positions in real time by using distributed computing and communication resources on the robots. The methods should first be developed and tested using numerical simulations and later validated on robot platforms under realistic operations.

  • Literature research in the field of autonomous swarm exploration, signal processing via networks and robotics, processing and evaluation of the state-of-the-art methods
  • Development of theoretically optimal procedures for autonomous swarm exploration of gases as well as procedures based, among others, on artificial intelligence methods and their combination with model-based approaches.
  • Theoretical and numerical analysis of the developed procedures
  • Comparison and analysis of the performance of newly developed methods with that of the already known approaches by means of KPIs (key performance indicators) such as speed of exploration and energy consumption of the swarm elements
  • Validation of the developed swarm exploration procedures in experiments
  • Adaptation of the developed procedures to computing modules used on experimental platforms
  • Integration of the algorithms on experimental platforms
  • Design, preparation and execution of experiments in indoor and outdoor environments with rovers and flying platforms
  • Evaluation of the results obtained in the experiments based on KPIs. Comparison with the theoretically achieved results and interpretation of differences between theory and experiments
  • Publication of the results in international journals and presentation of the results at professional conferences

Your qualifications:

  • Completed university degree (Master's/Diploma) in electrical engineering, communications engineering, computer science, mechanical engineering, or closely related scientific field
  • Expertise in image/signal processing, estimation and probability theory, or the willingness to acquire this expertise in a short period of time
  • Very good programming skills in, e.g. MATLAB, C++, Python
  • Basic knowledge of machine learning, physics and mathematics, systems integration, embedded systems.
  • Proficiency in written and spoken English
  • Ability to work independently as well as in a team-oriented, collaborative manner

Your benefits:

Look forward to a fulfilling job with an employer who appreciates your commitment and supports your personal and professional development. Our unique infrastructure offers you a working environment in which you have unparalleled scope to develop your creative ideas and accomplish your professional objectives. Our human resources policy places great value on a healthy family and work-life-balance as well as equal opportunities for persons of all genders (f/m/x). Individuals with disabilities will be given preferential consideration in the event their qualifications are equivalent to those of other candidates.

  • Apply online now
  • You can send this job advertisement via e-mail and complete your application on a personal computer or laptop.

    We need your digital application documents (PDF). The document upload function is not supported by all mobile devices. Please complete your application on a PC/laptop.

    Complete application on PC

Technical contact

Dr. Dmitriy Shutin
Institute of Communications and Navigation

Phone: +49 8153 28-2873

Send message

Vacancy 60105

HR department Oberpfaffenhofen

Send message

DLR site Oberpfaffenhofen

To location

DLR Institute of Communications and Navigation

To institute