image: In the modelic control paradigm, the first step is to establish a dynamic model through system identification. This model offers a continuous but inaccurate description of state transition information. In the datatic control paradigm, data is used directly for system analysis and controller synthesis, providing a discrete yet precise description of state transition.
Credit: Communications in Transportation Research, Tsinghua University Press
Researchers at Tsinghua University introduced a new concept called ϵ-controllability and its testing method for datatic control (i.e., data-driven control) systems, overcoming the challenge of controllability test for datatic systems caused by spatial discontinuity of their system description.
They published their study on 4 November 2024, in Communications in Transportation Research.
Controllability is a fundamental property of control systems. How to test the controllability of a system is an important problem in control theory. Controllability test theory for modelic control (i.e., model-based control) systems has been extensively studied by many renowned scholars. The concept of controllability was first introduced by Rudolf Kalman, the creator of the Kalman filter, in the 1960s. Subsequently, three experts in control theory—Popov, Belevitch, and Hautus—jointly proposed the PBH test method for linear time-invariant systems, providing a unified theoretical framework for verifying system properties such as controllability. Although controllability test theories for modelic control systems are well established, similar theories for general datatic control systems have yet to be developed. This is because modelic controllability test relies on known dynamic models that can provide spatially continuous system description, while datatic control systems are only described by discrete data points.
“Our proposed concept of ϵ-controllability establishes a basic theory for controllability analysis of datatic control systems and lays a foundation for controller design in the field of datatic control”, says Professor Shengbo Eben Li from School of Vehicle and Mobility at Tsinghua University.
Extending exact controllability to ϵ-controllability
Traditional controllability requires the system to be steered from an initial state exactly to a target state. In datatic systems, verifying exact controllability is impractical because the system is only described by discrete data points, and the dynamics information between data points is unknown. The concept of ϵ-controllability introduced by the research group extends exact controllability from a point-to-point form to a point-to-region form. It concerns whether the initial state can be transferred to a small neighborhood of the target state rather than exactly at that state.
Testing ϵ-controllability with MECS algorithm
The research group proposed a tree search algorithm called MECS for testing ϵ-controllability of datatic control systems. MECS performs the tree search by iteratively executing four steps: selection, expansion, evaluation, and pruning, until all ϵ-controllable states in the dataset are found. Experiment results on three datatic control systems show that the MECS algorithm can accurately test the ϵ-controllability of both linear and nonlinear datatic systems, successfully identifying all ϵ-controllable states in given datasets.
Future research can further explore how to reduce the time complexity of datatic controllability test algorithm, and how to test controllability of datatic systems with dynamic disturbance and observation noise.
The above research is published in Communications in Transportation Research (COMMTR), which is a fully open access journal co-published by Tsinghua University Press and Elsevier. COMMTR publishes peer-reviewed high-quality research representing important advances of significance to emerging transport systems. COMMTR is also among the first transportation journals to make the Replication Package mandatory to facilitate researchers, practitioners, and the general public in understanding and advancing existing knowledge. At its discretion, Tsinghua University Press will pay the open access fee for all published papers from 2021 to 2025.
About Communications in Transportation Research
Communications in Transportation Research was launched in 2021, with academic support provided by Tsinghua University and China Intelligent Transportation Systems Association. The Editors-in-Chief are Professor Xiaobo Qu, a member of the Academia Europaea from Tsinghua University and Professor Shuai’an Wang from Hong Kong Polytechnic University. The journal mainly publishes high-quality, original research and review articles that are of significant importance to emerging transportation systems, aiming to become an international platform and window for showcasing and exchanging innovative achievements in transportation and related fields, to promote the exchange and development of transportation research between China and the international academic community. It has been indexed in ESCI, Ei Compendex, Scopus, DOAJ, TRID and other databases. In 2022, it was selected as a high-starting-point new journal project of the “China Science and Technology Journal Excellence Action Plan”. This year, it received the first impact factor of 12.5. The 2023 IF is 12.5, ranking in the Top1 (1/57, Q1) among all journals in "TRANSPORTATION" category. At its discretion, Tsinghua University Press will pay the open access fee for all published papers from 2024 to 2025.
About Tsinghua University Press
Established in 1980, as a department of Tsinghua University, Tsinghua University Press (TUP) is a leading comprehensive higher education and professional publisher in China. TUP publishes 59 journals and 41 of them are in English. There are 16 journals indexed by SCIE/ESCI. Three of them have the highest impact factor in its field. In 2022, TUP launched SciOpen. As a publishing platform of TUP, SciOpen provides free access to an online collection of journals across diverse academic disciplines and serves to meet the research needs of scientific communities. SciOpen provides end-to-end services across manuscript submission, peer review, content hosting, analytics, identity management, and expert advice to ensure each journal’s development.
Journal
Communications in Transportation Research
Article Title
Controllability test for nonlinear datatic systems
Article Publication Date
4-Nov-2024