Knowledge Center

Assessment of State-of-Charge Estimation Method for Lithium-Ion Batteries

In this paper, a numerical model of lithium-ion batteries is developed and deployed to a Speedgoat Baseline target machine. The estimation method for the state-of-charge (SOC), based on a nonlinear autoregressive with exogenous input (NARX) and artificial neural networks (ANNs) that are correctly trained with multiple datasets, is designed, and experimentally validated by hardware-in-the-loop simulation.

Publication on mdpi.com

Hardware-in-the-Loop Testing (HIL) of State-of-Charge (SoC) Estimation for Li-Ion Batteries

This study presents the design and validation of an SoC estimation method for lithium-ion batteries in hybrid-electric vehicles (HEV). The battery model is deployed on a Speedgoat Baseline machine connected to a Raspberry Pi emulating the ECU based on an artificial neural network for HIL testing. The algorithm can estimate the SoC of the battery with 2% accuracy during real-time testing.

Independent Generation of Sequence Elements by Motor Cortex

Rapid execution of motor sequences depends on fusing movement elements into cohesive units that are executed holistically. The contribution of the primary motor and dorsal premotor cortex to this ability is determined in this paper. Also, the hypothesis that movement elements fuse makes specific predictions regarding three forms of activity, preparation, initiation, and execution is investigated.

Publication on nature.com

Motion Planning and Experimental Validation for an Autonomous Bicycle

This paper introduces a prototype autonomous two-wheeled vehicle developed for experimental verification of motion planning and control algorithms. Finally, it presents and discusses experiments run on the actual vehicle for a particular maneuver. It emphasizes the differences between the trajectories created by different vehicle models.

Publication on ieeexplore.ieee.org

A Software Architecture for an Autonomous Racecar

The authors present a detailed description of the software architecture used in the autonomous Roborace vehicles by the TUM-Team. The architecture combines the autonomous software functions perception, planning, and control, which are modularized for use on different hardware and to drive the car on high-speed racetracks. 

Publication on ieeexplore.ieee.org

Minimum Curvature Trajectory Planning and Control for an Autonomous Race Car

This paper shows a software stack capable of planning a minimum curvature trajectory for an autonomous race car based on an occupancy grid map. It introduces a controller design that allows following the trajectory at the handling limits. The the quadratic optimization problem is extended  for improved accuracy, the introduction of curvature constraints, and the reduction of linearization errors in corners.

Publication on tandfonline.com

Systematic Design of Multivariable Fuel Injection Controllers for Advanced Diesel Combustion

With multiple fuel injections per combustion cycle, the advanced diesel combustion process depends on all injection pulses in a coupled way. This work describes the cycle-to-cycle fuel injection control problem. A control-oriented model is introduced and locally validated with experimental data. Finally, a systematic design approach is proposed to synthesize a multivariable fuel injection controller.

Publication on ieeexplore.ieee.org

Control for a Supercapacitor Hybrid Energy Storage System Used in Electric Vehicles

This paper proposes the control strategy of a fully active hybrid energy storage system, which uses two bi-directional DC/DC converters to decouple supercapacitors and the battery pack from the DC bus. A Lyapunov-function-based controller regulates DC bus voltage, and a sliding mode controller, controlling the battery and supercapacitor currents, is designed. Their performance is validated by simulation and experimental data. 

Publication on sciencedirect.com