NM

M. Sc. Nina Majer

  • FZI Forschungszentrum Informatik
    Embedded Systems and Sensors Engineering (ESS)
    Haid-und-Neu Str. 10-14
    76131 Karlsruhe

Curriculum Vitae

Studies of Mechatronics and Information Technology at the Karlsruhe Institute of Technology (KIT) and the Pennsylvania State University (PSU) with the field of specification in Control in Mechatronics. 2020 Master's thesis "Safely Learning Predictive Adaptive Cruise Control for Highly Automated Vehicles“ at the FZI.                                                                                                                                       
Since April 2021 Research scientist at the department Control in Information Technology (CIT) in the research division Embedded Systems and Sensors Engineering (ESS) at the FZI.

Research

Cooperative Trajectory Planning in Mobile Robotics

The motion of multiple mobile robots in the same operation area can lead to intersection situations that cannot be resolved by individual and uncoordinated trajectory planners. To establish a coordinated planning method, centralized and decentralized control architectures have been proposed. Centralized architectures require a central coordination unit that is responsible for transmitting trajectories or priorities to the robots. In decentralized approaches, trajectory planning is equally distributed among the robots and explicit or implicit communication between the vehicles is necessary to resolve the conflict.
The trajectory planning problems of the mobile robots in an intersection scenario can be formulated as coupled optimal control problems, representing a differential game in the context of game theory. The proposed decentralized game-theoretic trajectory planning techniques for interacting mobile robots are mainly based on non-cooperative game theory. In non-cooperative game theory, each player minimizes its own cost function and does not consider the negative influence of its chosen strategy on other players’ costs. As a result, social effects when multiple robots interact are not modelled.
Therefore, a decentralized and cooperative trajectory planner to efficiently resolve intersection scenarios without the dependency on a central coordination unit shall be developed. A preferably small percentage of explicit communication between the robots should be required.

Teaching

Publications


Distributed model predictive control for cooperative autonomous lane merging
Gallant, M.; Schmidt, K.; Reimann, S.; Berkel, F.; Majer, N.; Hohmann, S.
2024. 2024 European Control Conference (ECC), Stockholm, 25th-28th June 2024, 2704 – 2711, Institute of Electrical and Electronics Engineers (IEEE). doi:10.23919/ECC64448.2024.10591087
Game-Theoretic Trajectory Planning of Mobile Robots in Unstructured Intersection Scenarios
Majer, N.; Luithle, L.; Schürmann, T.; Schwab, S.; Hohmann, S.
2023. IFAC-PapersOnLine, 56 (2), 11808–11814. doi:10.1016/j.ifacol.2023.10.575
Safely Learning Model Predictive Control with Time-Variant State Constraints and its Application to Motion Planning
Kohrer, L.; Majer, N.; Schwab, S.; Hohmann, S.
2021. 2021 IEEE International Intelligent Transportation Systems Conference (ITSC): 19-22 September 2021, Indianapolis, IN, USA, 755–762, Institute of Electrical and Electronics Engineers (IEEE). doi:10.1109/ITSC48978.2021.9564934