ACCOST
Abstract
Decades of successful applications have hardly exhausted the potential and vitality of the control technology. On the contrary, new societal imperatives highlight the importance of control, and investments in control technology are taking place in old and new industrial sectors. However, still the answer to the fundamental question: “Is deep control expertise nowadays sufficient to make a significant impact on industry and society?” is an emphatic NO.
New developments in mathematical systems, control theory and algorithms are needed to meet the challenges, with the focus mainly lying on issues such as robustness, stability and adaptability, and partially on cross-disciplinary endeavors in areas such as complexity, real-time systems and implementation. Towards this direction, the ACCOST project aims at: (i) proposing a consolidated methodology for nonlinear control systems focusing mainly on the issues of robustness and performance in conjunction with the operational constraints that arise either from the actuation level or the sensing and (ii) addressing issues and requirements involved in realizing practical and successful deployments.
For this reason, multiple experimental and realistic simulation studies will be conducted for a variety of application domains in aerial robotics and certain scientific/technological aspects that are critical to understand, while attempting to achieve impact with advanced control, will be examined. In this spirit, the envisioned unifying approach will yield a completely automated control system, which will pave the way towards actual autonomy in various control engineering domains. Notice that the consideration of multiple and probably coupled/conflicting operational constraints as well as their combination towards a unifying control approach call for new ways of thinking and analysis, which render the ACCOST project a beyond the state of the art and ground breaking approach to the field of nonlinear control systems engineering.
Objectives
Objective 1: Unsupervised Nonlinear System Identification
Within the ACCOST** project, motivated by previous work on prescribed performance control, we aim at designing: (i) an on-line RBF neural network identification scheme; (ii) a PE reference trajectory and (iii) a controller generating the appropriate control input signals that guarantee the satisfaction of PE condition for the RBF regressors employed. As a consequence, the neural network weight estimates would be not only uniformly bounded but also would converge to small neighborhoods of their actual values; thus achieving the goal of learning the actual system nonlinearities with quality guarantees.
Objective 2: Control under Actuation Limitations
Within the ACCOST project and motivated by previous works on prescribed performance control with deadzone, backlash and hysteresis input nonlinearities, we aim at developing a novel adaptive modification of the Prescribed Performance Control methodology that adjusts online the performance specifications when the input constraints tend to be violated, i.e., when a contradiction between what is desired and what is feasible arises. In particular, the ultimate goal will be to find out how input constraints affect the output performance through the system dynamics, which still remains open in the related literature since no constructive method has ever been proposed that achieves the best feasible performance under certain actuation constraints.
Objective 3: Control under Feedback Limitations
Within the ACCOST project and motivated by previous works on prescribed performance control, we aim at designing a novel observer scheme, reinforced with predefined performance attributes without either resorting to the high-gain approach or incorporating a priori knowledge of the actual system nonlinearities. Therefore, the intriguing properties of Prescribed Performance Control methodology would be sufficient to establish output feedback control with prescribed performance via simply securing the boundedness of the closed-loop system trajectories, thus resulting in significantly less complex and more robust design, compared to existing works in the related literature.
Objective 4: Demonstration in Aerial Robotics
It is very important to demonstrate and validate the overall control framework that will be designed in the context of the ACCOST project. For this reason, we aim at multiple experimental and realistic simulation studies that will incorporate all aforementioned theoretical aspects for a variety of application domains in aerial robotics. In particular within ACCOST, we shall consider multi-copter aerial robots involved in three case-studies: (i) trajectory tracking in free space, (ii) contact force/position control (in such case the multi-copter will be enhanced with a manipulator for increased intervention capabilities) and (iii) multi-robot cooperative formation. The studies should involve: (a) realistic modeling of the complex dynamics (including external disturbances and measurement noise) in each case-study, (b) an autonomous system identification module to extract an accurate dynamic model, (c) a control scheme that takes into account actuation as well as feedback constraints and guarantees certain predefined performance requirements imposed by the operational specifications of each case separately, (d) an automatic tuning process for the control parameters and (e) extensive comparative results with the current state of the art algorithms that are employed in each case-study, in order to put forward the full prospect of the proposed control framework.
Workplan
The ACCOST project involves the integration of novel and groundbreaking results within the fields of nonlinear control systems, system identification and robotics. Therefore, we believe that it is reasonable to expect that the full extent of three years is required for the project fulfillment. The work of the ACCOST project is divided into five work packages.
WP1 concerns the management of the entire project and all dissemination activities (exploitation of results, publicity events, etc.). The research and development part of the work comprises 4 work packages, which correspond to the key objectives described earlier.
In particular, WP2 will provide the model information for the control design in WP3 and WP4, which is necessary to achieve increased levels of robustness and automatic control gain tuning.
WP3 and WP4 will interact through the dependency and co-design of input constrained and output feedback control approaches respectively.
The interactions among these three WPs will be implemented and will in turn be integrated and demonstrated on the concrete scenarios described in WP5. Along the duration of the project, the tentative milestones with respect to the particular objectives are given as follows.
Milestone 1 (M18): We shall provide initial results on nonlinear system identification as well as on nonlinear observer-based control. In particular, initial elements of WP2 and WP4 will be integrated and demonstrated on the basis of the scenarios defined in WP5.
Milestone 2 (M27): We shall have further results on the fundamental limitations of nonlinear system identification and nonlinear observer-based control. Moreover, initial results on input constrained control design will also be given. We envision that this milestone will include the first attempt towards fully integrating WP2-WP4 in the application scenarios of the project.
Milestone 3 (M33): We shall have derived the overall control framework to handle input and feedback constraints for nonlinear dynamical systems. In this milestone, we aim at establishing and demonstrating at full extend the capabilities of the proposed methodology, with the focus lying on the selected application domains.
Deliverables
| No | Name | WP | Type | Dissemination | Due date |
|---|---|---|---|---|---|
| D1.1 | Project website | 1 | DEC | PU | M2 |
| D1.2 | Project results archived and available | 1 | DEC | PU | M36 |
| D2.1 | Reference trajectory generator | 2 | R + DEC | PU | M12 |
| D2.2 | Nonlinear system identification | 2 | R + DEC | PU | M24 |
| D3.1 | Control design under magnitude constraints | 3 | R + DEC | PU | M27 |
| D3.2 | Control design under slew-rate constraints | 3 | R + DEC | PU | M33 |
| D4.1 | Design of nonlinear observers | 4 | R + DEC | PU | M24 |
| D4.2 | Output feedback control design | 4 | R + DEC | PU | M30 |
| D5.1 | Scenarios Definition | 5 | R + DEC | PU | M12 |
| D5.2 | Milestone Demonstration 1 | 5 | R + DEM + DEC | PU | M18 |
| D5.3 | Milestone Demonstration 2 | 5 | R + DEM + DEC | PU | M27 |
| D5.4 | Milestone Demonstration 3 | 5 | R + DEM + DEC | PU | M33 |
| D5.5 | Assessment and Conclusions | 5 | R + DEC | PU | M36 |
Publications
- P. S. Trakas, C. P. Bechlioulis, “Approximation-free Adaptive Prescribed Performance Control for Unknown SISO Nonlinear Systems with Input Saturation”, in 2022 IEEE Conference on Decision and Control, Cancun, Mexico, Dec 6 – 9, 2022.
- S. Tzeranis, P. S. Trakas, X. Papageorgiou, K. J. Kyriakopoulos and C. P. Bechlioulis, “Robust Prescribed Performance Control and Adaptive Learning for the Longitudinal Dynamics of Fixed-Wing UAVs”, in 2022 International Conference on Systems and Control (ICSC), Marseille, France, Nov 24 – 25, 2022.
- A. Samanis, P. S. Trakas, X. Papageorgiou, K. J. Kyriakopoulos and C. P. Bechlioulis, “Learning Unknown Lagrange Dynamical Systems with Guaranteed Persistency of Excitation”, in 2022 International Conference on Systems and Control (ICSC), Marseille, France, , Nov 24 – 25, 2022.
- P. S. Trakas, C. P. Bechlioulis, “Robust Adaptive Prescribed Performance Control for Unknown Nonlinear Systems With Input Amplitude and Rate Constraints,” in IEEE Control Systems Letters, vol. 7, pp. 1801-1806, 2023, doi: 10.1109/LCSYS.2023.3281346.
- P. S. Trakas, C. P. Bechlioulis, “Robust Trajectory Tracking Control for Constrained Small Fixed-Wing Aerial Vehicles with Adaptive Prescribed Performance”, MDPI Applied Sciences, 2023, 13, 7718; https://doi.org/10.3390/s23010221.
The project is funded by Hellenic Foundation for Research and Innovation (H.F.R.I.) under the “2nd Call for H.F.R.I. Research Projects to support Post-Doctoral Researchers” (project number 370).
People
Charalampos Bechlioulis
Panagiotis Trakas
Christos Vlachos
