dtControl: Decision Tree Learning Algorithms for Controller Representation (Python toolbox)

Last updated on Mar 7, 2020

The dtControl provides the memory-efficient, compact, understandable representation and the efficient determinization for the formal symbolic controllers obtained from the state-of-the-art toolboxes SCOTS and UPPAL. It also provides a scheme to design non-uniform quantizers (i.e., state encoders with non-uniform partitioning of state-set) for symbolic controllers. The results are also useful for constructing efficient static coders minimizing bit rate over the sensor-controller channel.

Download dtControl

Install using pip package click here

  1. User Manual
  2. P. Ashok, M. Jackermeier, P. Jagtap, J. Kretinsky, M. Weininger, M. Zamani (2020). dtControl: Decision tree learning algorithms for controller representation. 23rd International Conference on Hybrid Systems: Computation and Control (HSCC).
Avatar
Pushpak Jagtap
Assistant Professor