Under complex and uneven road conditions, variations in vehicle load can readily induce stochastic disturbances in key parameters such as the center of mass and moments of inertia. These numerous and unpredictable fluctuations introduce stochastic structural uncertainties into the full-vehicle dynamic model, potentially destabilizing conventional suspension control strategies and degrading their overall performance. To address these challenges, this paper investigates a data-driven reinforcement learning-based H2/H∞ vibration control strategy, aiming to achieve robust suspension control in the presence of stochastic structural uncertainties. Firstly, a control-oriented model of the full-vehicle active suspension system is constructed. The control problem is reformulated as a stochastic zero-sum game problem, and an iterative algorithm is given to solve the associated stochastic game algebraic Riccati equation. To alleviate the complexity and high cost of parameter tuning and model identification, a data-driven off-policy reinforcement learning algorithm is introduced. This approach effectively compensates for the adverse effects introduced by stochastic structural uncertainties and provides a robust control solution without requiring explicit knowledge of system parameters. Finally, numerical simulations are conducted to validate the robustness of the proposed approach. The results show that the proposed algorithm can converge to the optimal solution within 10 steps.
LiuZSiYSunW.Ride comfort oriented integrated design of preview active suspension control and longitudinal velocity planning. Mech Syst Signal Process2024; 208(15): 110992.
2.
SinghAK.Design of robust active suspension system for performance improvement and vibration control of railway vehicles: an H∞ and μ-synthesis perspective. Proc Inst Mech Eng Part C: J Mech Eng Sci2024; 238(22): 10609–10631.
3.
ZhuJZhaoDChenH, et al. Nonlinear force transfer characteristic analysis of active suspension for emergency rescue vehicles. Proc Inst Mech Eng Part C: J Mech Eng Sci2025; 239(7): 2667–2677.
4.
PanHSunWGaoH, et al. Nonlinear tracking control based on extended state observer for vehicle active suspensions with performance constraints. Mechatronics2015; 30: 363–370.
5.
RiccoMAlshawiAGruberP, et al. Nonlinear model predictive control for yaw rate and body motion control through semi-active and active suspensions. Veh Syst Dyn2024; 62(6): 1587–1620.
6.
PanHZhangCSunW.Fault-tolerant multiplayer tracking control for autonomous vehicle via model-free adaptive dynamic programming. IEEE Trans Reliab2023; 72(4): 1395–1406.
7.
KimballJBDeBoerBBubbarK.Adaptive control and reinforcement learning for vehicle suspension control: a review. Annu Rev Control2024; 58: 100974.
8.
WangBSunJPengB, et al. Optimal event-triggered neural learning tracking control for pneumatic muscle antagonistic joint with asymmetric constraints. IEEE Trans Ind Electron2025; 72: 14677–14687.
9.
HaddarMChaariRBaslamisliSC, et al. Intelligent PD controller design for active suspension system based on robust model-free control strategy. Proc Inst Mech Eng Part C: J Mech Eng Sci2019; 233(14): 4863–4880.
10.
DingRWangRMengX, et al. Intelligent switching control of hybrid electromagnetic active suspension based on road identification. Mech Syst Signal Process2021; 152(1): 107355.
11.
AfsharKKKorzeniowskiRKoniecznyJ.Constrained dynamic output-feedback robust H∞ control of active inerter-based half-car suspension system with parameter uncertainties. IEEE Access2023; 11: 46051–46072.
12.
YenenehKWalleMMamoT, et al. Optimizing active suspension systems with robust H∞ control and adaptive techniques under uncertainties. Appl Eng Sci2025; 22: 100225.
13.
ZhangMJingXWangG.Bioinspired nonlinear dynamics-based adaptive neural network control for vehicle suspension systems with uncertain/unknown dynamics and input delay. IEEE Trans Ind Electron2021; 68(12): 12646–12656.
14.
XuYXieZZhaoJ, et al. Robust non-fragile finite frequency H∞ control for uncertain active suspension systems with time-delay using T-S fuzzy approach. J Franklin Inst2021; 358(8): 4209–4238.
15.
LiuLLiXLiuYJ, et al. Neural network based adaptive event trigger control for a class of electromagnetic suspension systems. Control Eng Pract2021; 106: 104675.
16.
ZhouZZhangMNavarro-AlarconD, et al. Model-free finite-time saturated control for active vehicle suspension systems with dead zones and external disturbances. Mech Syst Signal Process2025; 229(15): 112542.
17.
HanSYZhouJChenYH, et al. Active fault-tolerant control for discrete vehicle active suspension via reduced-order observer. IEEE Trans Syst Man Cybern Syst2021; 51(11): 6701–6711.
18.
LiZChenHWuW, et al. Dynamic output feedback fault-tolerant control for switched vehicle active suspension delayed systems. IEEE Trans Vehicular Technol2024; 73(8): 11059–11071.
19.
XuXLiB.Semi-global stabilization of parabolic PDE–ODE systems with input saturation. Automatica2025; 171: 111931.
20.
SunWZhaoYLiJ.Active suspension control with frequency band constraints and actuator input delay. IEEE Trans Ind Electron2012; 59(1): 530–537.
21.
XueWLiKChenQ, et al. Mixed fts/H∞ control of vehicle active suspensions with shock road disturbance. Veh Syst Dyn2019; 57(6): 841–854.
22.
LiWDuXXiaD, et al. Robust H∞ control for magnetorheological suspension system of all-terrain vehicles with parameter uncertainties and time delay. Veh Syst Dyn Epub ahead of print 28 May 2025. DOI: 10.1080/00423114.2025.2501993.
23.
ChenGDuGXiaJ, et al. A periodic sampled-data H∞ control of vehicle active suspension system: an uncertain discrete-time model approach. IEEE Trans Ind Inform2024; 20(4): 6739–6750.
24.
ShiZCaoRZhangS, et al. Active suspension H∞/generalized H2 static output feedback control. J Vib Control2024; 30(23-24): 5183–5195.
25.
JiangYXieSChenG.Data-driven H∞ control of networked nonlinear systems with external disturbances and random communication packet losses. IEEE Trans Control Netw Syst2024; 11(3): 1358–1369.
26.
YinYLuoSDengF.Stochastic H2/H∞ off-policy reinforcement learning tracking control for linear discrete-time systems with multiplicative noises. J Franklin Inst2025; 362(1): 107349.
27.
QinZCXinY.Data-driven H∞ vibration control design and verification for an active suspension system with unknown pseudo-drift dynamics. Commun Nonlinear Sci Numer Simul2023; 125: 107397.
28.
WangCTaoGCuiX, et al. Mechanism-data-driven control strategy for active suspension systems: integrating deep reinforcement learning with differential geometry to enhance vehicle ride comfort. Adv Eng Inform2025; 65: 103326.
29.
WangGLiKLiuS, et al. Model-free H∞ output feedback control of road sensing in vehicle active suspension based on reinforcement learning. J Dyn Syst Meas Control2023; 145(6): 1–12.
30.
WangGDengJDuanD, et al. Constrained H∞ optimal control for nonlinear active suspensions via data-driven reinforcement learning algorithm. J Comput Nonlinear Dyn2025; 20(7): 071007.
31.
GaoXDuYHanS, et al. Multi-agent reinforcement learning for vibration control of regenerative active suspension. Eng Appl Artif Intell2025; 153: 110809.
32.
MazouchiMYangYModaresH.Data-driven dynamic multiobjective optimal control: an aspiration-satisfying reinforcement learning approach. IEEE Trans Neural Netw Learn Syst2022; 33(11): 6183–6193.
33.
WangCCuiXZhaoS, et al. A deep reinforcement learning-based active suspension control algorithm considering deterministic experience tracing for autonomous vehicle. Appl Soft Comput2024; 153: 111259.
34.
ZhaoYZhangYGuoL, et al. Advances in machine learning-based active vibration control for automotive seat suspensions: a comprehensive review. Mech Syst Signal Process2025; 231(15): 112645.
35.
ZhangXLiuYChenX, et al. Adaptive pseudoinverse control for constrained hysteretic nonlinear systems and its application on dielectric elastomer actuator. IEEE/ASME Trans Mechatron2023; 28(4): 2142–2154.
36.
ShenHZhaoWCaoJ, et al. Predefined-time event-triggered tracking control for nonlinear servo systems: a fuzzy weight-based reinforcement learning scheme. IEEE Trans Fuzzy Syst2024; 32(8): 4557–4569.
37.
ShenHPengCYanH, et al. Data-driven near optimization for fast sampling singularly perturbed systems. IEEE Trans Automat Contr2024; 69(7): 4689–4694.
38.
ZhouZWangYLiuX, et al. Hybrid of neural network and physics-based estimator for vehicle longitudinal dynamics modeling using limited driving data. IEEE Trans Intell Transp Syst2025; 26: 16735–16746.
39.
JiangYGaoWChenC, et al. Adaptive optimal control of linear discrete-time networked control systems with two-channel stochastic dropouts. SIAM J Control Optim2023; 61(5): 3183–3208.
40.
LaiJXiongJShuZ.Model-free optimal control of discrete-time systems with additive and multiplicative noises. Automatica2023; 147: 110685.
41.
LiNLiXPengJ, et al. Stochastic linear quadratic optimal control problem: a reinforcement learning method. IEEE Trans Automat Contr2022; 67(9): 5009–5016.
42.
ZhouZWangYZhouG, et al. Vehicle lateral dynamics-inspired hybrid model using neural network for parameter identification and error characterization. IEEE Trans Vehicular Technol2024; 73(11): 16173–16186.
43.
ZhangWXieLChenB.Stochastic H2/H∞ control: a nash game approach. CRC Press, 2017.
44.
BekirosSJahanshahiHBezzinaF, et al. A novel fuzzy mixed H2/H∞ optimal controller for hyperchaotic financial systems. Chaos Solitons Fractals2021; 146: 110878.
45.
ZhangWHuangYXieL.Infinite horizon stochastic H2/H∞ control for discrete-time systems with state and disturbance dependent noise. Automatica2008; 44(9): 2306–2316.
46.
KoralovLSinaiYG.Theory of probability and random processes. Springer Science & Business Media, 2007.
