Jul 2026· Proceedings of the Genetic and Evolutionary Computation Conference Companion· pp. 609-612· 0 citations· 9 references
TL;DR
This paper presents a hardware-in-the-loop approach that optimizes PI coefficients directly on a physical machine, eliminating the need for transfer function identification and simulation.
Abstract
Determining suitable PI controller coefficients remains a labor-intensive process that often relies on trial-and-error or analytical methods requiring detailed system models. This paper presents a hardware-in-the-loop approach that optimizes PI coefficients directly on a physical machine, eliminating the need for transfer function identification and simulation. The machine operates as a black box, with candidate solutions evaluated through measured step responses via a remote procedure call interface. Three evolutionary algorithms, Differential Evolution (DE), Particle Swarm Optimization (PSO), and Covariance Matrix Adaptation Evolution Strategy (CMA-ES), were compared under strict real-world constraints: one-minute evaluation times and a total optimization budget of one hour. Initial experiments optimizing only the velocity controller achieved up to 90% improvement but introduced oscillation artifacts caused by an inverse dependency between the cascaded velocity and torque controllers. Simultaneous optimization of both controllers resolved these side effects, yielding a 48% reduction in overshoot and 43% in undershoot compared to the manufacturer's default parameters. All algorithms converged within approximately ten minutes. DE and PSO are recommended as the most effective choices for this optimization scenario.
The primary objective of this protocol is to provide a reproducible fixed-step simulation framework for evaluating Particle Swarm Optimization (PSO)-based gain tuning in nonlinear control systems. The implementation begins with the formulation of a Furuta-type pendulum model, followed by the integration of a backstepping controller within a 2 ms fixed-step execution environment. The methodology involves a systematic four-stage process: characterizing non-ideal implementation constraints, defining a multi-objective Particle Swarm Optimization search space, executing automated offline tuning, and evaluating the resulting parameters through a standardized suite of trajectory-tracking and disturbance-rejection scenarios. This setup, utilizing high-performance industrial workstations and standardized signal interfaces, supports consistent repeated-trial comparisons within the same control architecture. The design compares a baseline manually tuned backstepping controller with a PSO-optimized variant that shares the exact same control structure, thereby isolating the impact of gain selection. Controller performance is assessed across three distinct operational scenarios: step-trajectory following, mixed-frequency sinusoidal tracking, and disturbance rejection. Statistical analysis of 10 repeated trials showed that PSO-based optimization reduced the step-tracking RMSE from 0.065 to 0.050 rad and attenuated peak pendulum excursions by 33.1%. These improvements were achieved alongside a 22.1% reduction in RMS control effort, indicating that the optimized parameters facilitated more efficient energy distribution within the Lyapunov-based framework. Ultimately, this methodology provides a structured simulation framework to evaluate nonlinear control strategies before any subsequent physical hardware implementation.
This paper proposes a synthesis procedure for a two-degree-of-freedom (2DOF) robust loop-shaping controller using the Particle Swarm Optimization (PSO) algorithm. The controller is designed to satisfy predefined performance constraints in both time and frequency domains. А key idea is the use of a generalized weighting function instead of a simple first-order one. This provides a more flexible framework for controller design. Coefficients of this generalized function are treated as decision variables in a constrained optimization problem. By integrating strengths of 2DOF robust loop shaping with computational capabilities of PSO in a closed-loop design, the synthesized controller demonstrates high robustness to significant system parameter variations. In addition, it helps reduce overshoot, shorten settling time, and improve disturbance rejection. Simulation results from a quadrotor attitude control system indicate that the proposed controller can provide robust performance under complex flight conditions. Moreover, its performance and robustness are validated through comparison with a controller designed using the NOMAD (Nonlinear Optimization with Mesh Adaptive Direct search)
K. B. Son, V. Tien, C. H. Tinh· Mekhatronika Avtomatizatsiya...· 0 citations
To address the difficulty of conventional fixed-parameter PI controllers in the inner current loop of Vienna rectifiers in simultaneously achieving a fast dynamic response and overshoot suppression, this paper proposes a PSO-enhanced BPNN-based PI parameter-tuning strategy (PSO-BPNN-PI). The global search capability of particle swarm optimization (PSO) is exploited to optimize the initial weights of the back-propagation neural network (BPNN), mitigating the local-minima problem of random initialization and yielding a better-performing set of inner-loop PI parameters. On the equivalent current-loop model, the proposed strategy reduces the step-response overshoot to only 3.35%. On a full switching model of the Vienna rectifier, the proposed strategy is evaluated under steady-state operation, load disturbance, and filter-inductance variation and is further benchmarked against conventional PI, randomly initialized BPNN-PI, and LADRC controllers in dynamic disturbance tests. The proposed method achieves the lowest input-current THD and a balanced overall performance, while remaining competitive in tracking accuracy and maintaining its advantage over a wide inductance-mismatch range. Finally, a Hardware-in-the-Loop (HIL) platform with a heterogeneous CPU/FPGA architecture is established, on which the offline-optimized PI parameters are deployed; the results show stable tracking of the 750 V DC-bus voltage, rapid recovery from sudden load steps, and a well-balanced neutral-point voltage, confirming the engineering feasibility of the offline-optimized parameters.
Jun Chen, Siqi Cheng, Ziyi Bai· Energies· 0 citations
This paper addresses experimental PID tuning for DC motor actuators when an accurate plant model is unavailable or impractical to obtain. Controller tuning is formulated as a constrained closed-loop optimization process in which each PID gain set is deployed on the physical system, produces an experimental dataset, and is evaluated through a performance index. The objective function combines tracking error, control effort, and control-signal variation, while penalty terms identify and penalize actuator saturation, excessive overshoot, settling-time violations, steady-state error, and divergent responses. Candidate controllers are generated using a constrained Gaussian Cross-Entropy Method and evaluated directly on the physical actuator. The proposed host-embedded architecture separates two computational time scales: PID execution, encoder processing, and data acquisition are performed in real time on the embedded platform, whereas population sampling, candidate ranking, and distribution updates are executed on the host computer between experiments. Within the reported experimental campaign, the sampling distribution progressively concentrates toward gain regions associated with lower closed-loop cost under the prescribed admissibility criteria. The resulting framework provides a structured and traceable procedure for physical controller deployment, data acquisition, constrained performance evaluation, and episodic PID retuning without explicit plant identification. The reported results demonstrate the feasibility of the architecture on the considered platform, without implying comparative superiority, run-to-run statistical convergence, or analytical closed-loop stability.
Jorge A. Lizarraga, L. Luque-Vega, Javier Ruiz-León et al.· Engineer· 0 citations
This paper presents a comprehensive experimental and simulation-based investigation of intelligent optimization techniques for PID-controlled AC servo drive systems. Three optimization approaches, namely the Genetic Algorithm (GA), Grey Wolf Optimization (GWO), and a Hybrid Genetic Algorithm Grey Wolf Optimization (HGAGWO), are employed to determine the optimal PID controller parameters for high-precision speed and position control under varying operating conditions and load disturbances. The proposed framework combines advanced optimization techniques with experimental validation to provide a reliable assessment of controller performance in practical industrial environments. The experimental platform consists of a programmable logic controller (PLC), a human-machine interface (HMI), an AC servo drive, and a high-resolution external encoder, while MATLAB/Simulink is utilized to develop and validate the dynamic model. Controller performance is systematically evaluated using widely accepted dynamic performance indices, including rising time, settling time, overshoot, steady-state error, and tracking capability under different operating scenarios. The controller optimization study addresses backlash, one of the most significant nonlinearities affecting servo drive accuracy. A backlash compensation strategy is implemented and experimentally verified, demonstrating a noticeable improvement in positioning precision and motion stability. Comparative simulation and experimental results confirm that the optimized PID controllers significantly enhance the transient response, disturbance-rejection capability, tracking accuracy, and overall system robustness compared with conventional PID tuning methods. Among the investigated optimization techniques, the proposed HGAGWO algorithm effectively combines the global exploration capability of GA with the fast convergence characteristics of GWO, producing superior optimization accuracy, faster convergence, and more reliable controller tuning. The close agreement between simulation and experimental results further validates the effectiveness and practical applicability of the proposed methodology. Therefore, the integration of intelligent PID optimization with backlash compensation provides a robust and efficient motion control solution for high-performance industrial servo systems operating under nonlinearities, load variations, and parameter uncertainties.
Shereen A. Fayad, M. Shaban, Mohamed I Attia et al.· Scientific Reports· 0 citations
Nonlinear underactuated systems pose significant challenges for controller design owing to their instability, nonlinear dynamics, strong state coupling, and sensitivity to parameter variations. This paper presents an adaptive hybrid optimization framework for controller-energy CODESIGN in nonlinear systems. The term ‘adaptive’ refers to the runtime adjustment of the search strategy rather than to an adaptive control law: the FOPID controller structure and parameters are fixed during closed-loop operation. The framework consists of two phases: a calibration stage focused on control accuracy and a CODESIGN stage that incorporates an energy-related term through a weighted objective. A conditional intensification mechanism supports the search under stagnation or near the end of the optimization process. Unlike simple sequential hybridization, the proposed Hybrid AHA-SMA (HAHASMA) activates the Slime Mould Algorithm (SMA) only when additional local refinement is required, whereas the Artificial Hummingbird Algorithm (AHA) remains the main global search engine. The method is evaluated on a cart-inverted pendulum system with a fractional-order proportional-integral-derivative (FOPID) controller under external disturbance and stochastic input noise, using a fixed function-evaluation budget and a same-start initialization protocol. Across 20 benchmark functions, HAHASMA obtained the best Friedman average rank (1.35). The calibration stage yielded the lowest run-to-run variability (std: 0.240 versus 30.574–164.522). In the CODESIGN stage, increasing $\beta $ from 0.00 to 0.60 reduced the composite objective from about 1.00 to 0.710 while the normalized control-quality constraint remained within the prescribed tolerance.
Merve Türktam, Hasan Zorlu, Selim Soylu· IEEE Access· 0 citations