Flexible pneumatic joint actuators are widely employed in medical, wearable device, and soft gripper applications due to their excellent adaptability. However, their joint angles are prone to oscillation due to material hyper elasticity and gas compressibility, thereby limiting control precision. To address this, this paper proposes a precise joint angle control method based on the CNN-AUPI (Convolutional Neural Network-Amplitude-dependent Un-parallel Prandtl-Ishlinskii) model. This approach employs the CNN-AUPI model to perform real-time fitting of the actuator’s output angle. It integrates the ISI (Iterative Search Inversion) algorithm for model inversion, enabling online adjustment of control pressure to compensate for angular errors. Experimental results demonstrate that this method achieves high-precision tracking of the joint along the desired trajectory, with a maximum mean absolute error of 0.28° under step inputs and a mean relative error not exceeding 1.59%. Compared with PI (Prandtl-Ishlinskii) and DRDPI (Deadband Rate-Dependent Prandtl-Ishlinskii) control methods, CNN-AUPI demonstrates superior performance in angular stability and control accuracy, offering an effective solution for stable angular control of flexible actuators.
LiGChenXZhouF, et al. Self-powered soft robot in the Mariana trench. Nature2021; 591: 66–71.
4.
GaiLJHuangJZongX. Stiffness-tunable soft bellows actuators by cross-fiber jamming effect for robust grasping. IEEE/ASME Trans Mechatron2023; 28: 2897–2907.
5.
WangJGaoDLeePS. Recent progress in artificial muscles for interactive soft robotics. Adv Mater2021; 33(19): 2003088.
6.
ZhaoHHussainAMIsrarA, et al. A wearable soft haptic communicator based on dielectric elastomer actuators. Soft Robot2020; 7: 451–461.
7.
ZhuYGongWChuK, et al. A novel wearable soft glove for hand rehabilitation and assistive grasping. Sensors2022; 22(16): 6294.
8.
LiuJWangXLiuS, et al. Vertebraic soft robotic joint design with twisting and antagonism. IEEE Robot Autom Lett2022; 7: 658–665.
9.
Banu SundareswariMThen MozhiGDhanalakshmiK. Angular control of differential shape-memory alloy spring actuator for underactuated dynamic system. J Vib Control2022; 28: 2829–2843.
10.
KimSOhYKangJ, et al. Design of an accordion-fold-inspired soft electrohydraulic actuator for angular motion. IEEE Robot Autom Lett2024; 9: 4750–4757.
11.
XiaoHMengQXLaiXZ, et al. Design and trajectory tracking control of a novel pneumatic bellows actuator. Nonlinear Dyn2023; 111: 3173–3190.
12.
HuangPWuJZhangP, et al. Dynamic modeling and tracking control for dielectric elastomer actuator with a model predictive controller. IEEE Trans Ind Electron2022; 69: 1819–1828.
13.
XieXCuiYYuY, et al. Modeling and identification of nonlinear hysteresis behavior of piezoelectric actuators using a computationally efficient phenomenological model and modified cuckoo search algorithm. Smart Mater Struct2023; 32(1): 015013.
14.
WangWWangJChenZ, et al. Research on asymmetric hysteresis modeling and compensation of piezoelectric actuators with PMPI model. Micromachines2020; 11(4): 357.
15.
LiangYGuanP. Improved Maxwell model with structural dashpot for characterization of ultraslow creep in concrete. Constr Build Mater2022; 329: 127181.
16.
LiDWangY. Parameter identification of a differentiable Bouc-Wen model using constrained extended Kalman filter. Struct Health Monit2021; 20: 360–378.
17.
ThaiMTPhanPTHoangTT, et al. Design, fabrication, and hysteresis modeling of soft microtubule artificial muscle (SMAM) for medical applications. IEEE Robot Autom Lett2021; 6: 5089–5096.
18.
SrivastavaAHuiCY. Large deformation contact mechanics of long rectangular membranes. I. Adhesion less contact. Proc R Soc A Math Phys Eng Sci2013; 469(2160): 20130424.
19.
LiZZhangXMaL. Development of a combined Prandtl Ishlinskii–Preisach model. Sens Actuators A Phys2020; 304: 111797.
20.
ZhangCZhouMNieL, et al. Prandtl–Ishlinskii model-based event-triggered prescribed control: design and application to piezoelectric-driven micro positioning stage. Mech Syst Signal Process2023; 200: 110562.
21.
SavoieMShanJ. Temperature-dependent asymmetric Prandtl-Ishlinskii hysteresis model for piezoelectric actuators. Smart Mater Struct2022; 31(5): 055022.
22.
YoongHSuCYeoK. Stress-dependent generalized Prandtl–Ishlinskii hysteresis model of a NiTi wire with super elastic behavior. J Intell Mater Syst Struct2021; 32: 1713–1724.
23.
WangWYaoJE. Modeling and identification of magneto strictive hysteresis with a modified rate-independent Prandtl–Ishlinskii model. Chin Phys B2018; 27(9): 098503.
24.
AbbasiPNekouiMAZareinejadM, et al. Position and force control of a soft pneumatic actuator. Soft Robot2020; 7: 550–563.
25.
ZhangYLiuHMaT, et al. A comprehensive dynamic model for pneumatic artificial muscles considering different input frequencies and mechanical loads. Mech Syst Signal Process2021; 148: 107133.
26.
ZhangYGaoJYangH, et al. A novel hysteresis modelling method with improved generalization capability for pneumatic artificial muscles. Smart Mater Struct2019; 28(10): 105014.
27.
NiuBLiuJDuanP, et al. Reduced-order observer-based adaptive fuzzy tracking control scheme of stochastic switched nonlinear systems. IEEE Trans Syst Man Cybern Syst2021; 51: 4566–4578.
28.
GouryODuriezC. Fast, generic, and reliable control and simulation of soft robots using model order reduction. IEEE Trans Robot2018; 34: 1565–1576.
29.
PengKChenWGuanS, et al. Hysteresis inversion-free predictive compensation control for soft pneumatic actuators based on a global Koopman modeling strategy. Phys Scr2023; 98(12): 125206.
30.
KimDKimJIParkYL. A simple tripod mobile robot using soft membrane vibration actuators. IEEE Robot Autom Lett2019; 4: 2289–2295.
31.
KimDKimSHKimT, et al. Review of machine learning methods in soft robotics. PLoS One2021; 16(2): e0246102.
32.
RegoRCBAraújoFMUD. Nonlinear control system with reinforcement learning and neural networks based Lyapunov functions. IEEE Lat Am Trans2021; 19(8): 1253–1260.
33.
LiuHChengQXiaoJ, et al. Data-driven adaptive integral terminal sliding mode control for uncertain SMA actuators with input saturation and prescribed performance. ISA Trans2022; 128: 624–632.
34.
WangMLeeWShuL, et al. Development and analysis of an origami-based elastomeric actuator and soft gripper control with machine learning and EMG sensors. Sensors2024; 24(6): 1751.
35.
ZhangATrubyRLChinL, et al. Vision-based sensing for electrically-driven soft actuators. IEEE Robot Autom Lett2022; 7: 11509–11516.
36.
HuJZhongYYangM. Hysteresis modeling of piezoelectric micro-positioning stage based on convolutional neural network. Proc Inst Mech Eng I J Syst Control Eng2021; 235: 170–179.
37.
WangF. Neural network model for hysteretic characteristic of shape memory alloy. Mater Today Commun2023; 35: 105963.
38.
KrikelisKPeiJSvan BerkelK, et al. Identification of structured nonlinear state–space models for hysteretic systems using neural network hysteresis operators. Measurement2024; 224: 113966.
39.
ChenSXuMLiuS, et al. CNN–AUPI-based force hysteresis modeling for soft joint actuator. Arab J Sci Eng2024; 49: 14577–14591.
40.
XuMWangGRongC. Fiber-reinforced flexible joint actuator for soft arthropod robots. Sens Actuators A Phys2022; 340: 113522.
41.
XiaoHLaiXMengQ, et al. Kinematic modeling and control for an elephant-trunk soft manipulator considering hysteresis. IEEE Trans Ind Inform2024; 20: 11318–11328.
42.
ShakibaSOurakMPoortenEV, et al. Modeling and compensation of asymmetric rate-dependent hysteresis of a miniature pneumatic artificial muscle-based catheter. Mech Syst Signal Process2021; 154: 107532.
