Antagonistic pneumatic muscle (PM)-actuated wrist robots have great potential in rehabilitation and industrial applications. The antagonistic connection of PMs, which mimics the agonist-antagonist muscle pairs in human joints, provides substantial advantages such as improved joint stability and a better balance of torque disturbances. However, PM-actuated robots exhibit complex nonlinearities, such as hysteresis, creep, input delay, and time-varying parameters, while also confronting challenges such as external disturbances and coupling effects. In this paper, a switching non-singular terminal sliding mode control (NTSMC) method with a double-loop fuzzy neural network (DLFNN) is developed. This method enables the antagonistic PM-actuated wrist robots to achieve fast and precise tracking performance. Specifically, the lumped disturbances are estimated online using the DLFNN, which can adaptively adjust the weight of the inner and outer layers, achieving accurate approximation and robustness. Based on the estimated value of disturbances, a switching NTSMC is implemented to ensure that tracking errors converge to the origin within the fixed time. Switching functions guarantee fast convergence when the sliding surface errors are large. Meanwhile, switching functions ensure non-singularity as the sliding surface errors converge to the origin. Furthermore, joint angles and angular velocities are limited within the specific ranges by designing exponential constraint terms as time-varying proportional-differential gains, rather than traditional barrier functions that may induce excessive control inputs. Both detailed stability analysis and experimental validation demonstrate the effectiveness and adaptability of the proposed method.

Pneumatic artificial muscle (PAM) actuators have passive compliance and vibration absorption capabilities, adapting to high-intensity human-robot interaction movements. Unfortunately, the asymmetric hysteresis of PAMs is prone to produce motion delays and control inaccuracy, and anti-disturbance control is not friendly when applied in exoskeleton robots. Also, most of existing studies on active compliant control are overly reliant on bulky force-sensing, which is limited by sampling accuracy and communication rate. Therefore, it is still a challenge to realize compliant motions of PAM systems, while ensuring the rapid convergence of output signals. To this end, a new practical finite-time compliant controller is designed in this paper, which realizes satisfactory tracking control of horizontal PAM systems. Specifically, the external contact force is estimated by an improved adaptive law, instead of sensor feedback, thus decreasing noise effects. Meanwhile, the proposed controller ensures that the output tracking error converges quickly within known finite time, while reducing computational complexity. In particular, the desired trajectory is updated in real time by a modified admittance model, so as to achieve motion compliance and interaction safety. Compared with the literature, it is the first attempt to provide a compliant control solution for horizontal PAM systems without force feedback information, and ensures the practical finite-time convergence of the output tracking error. The rigorous stability analysis is presented, and the reliability of the proposed method is verified by hardware experiments. Note to Practitioners—Owing to light weight and flexibility, pneumatic artificial muscle (PAM) actuators can better meet the growing demands of human-machine cooperation tasks. In areas such as power assist and rehabilitation training equipment, it is necessary to ensure that robots can properly adjust the motion trajectory according to the applied force, so as to replace unbearable “hard contact” with more dexterous “compliant interaction”. Inspired by this, a new compliant control method is designed in this paper for horizontal PAM systems, which realizes the online adjustment of desired motion trajectories according to contact forces, and ensures the rapid convergence of the tracking error within known finite time. In particular, the proposed method improves an efficient contact force estimation solution, avoiding bulky and expensive force-sensing, while reducing noise effects. Compared with existing results, this paper for the first time presents a compliant control method without sensor feedback of horizontal PAM systems, and improves the motion rapidity, accuracy, and stability in practical tracking and training cases. Meanwhile, rigorous stability analysis is provided by Lyapunov techniques, and the effectiveness of the proposed method is verified by experiments on a self-built PAM testbench. In the future, we will try to apply the proposed method to limb functional training scenarios, aiming to expand its practical prospects and increase efficiency.

As a kind of soft actuator with inherent compliance, pneumatic artificial muscles (PAMs) have great application potential in robots. However, some challenging issues, such as high nonlinearities, sensor noises, and external disturbances, inevitably bring enormous difficulties to the accurate control of PAM-driven robots. To this end, this paper proposes an adaptive output feedback fuzzy switching control method for switched-form PAM-driven parallel robot systems, utilizing admittance models to rebuild compliant trajectories. Specifically, based on the nonrecursive high-order sliding mode (HOSM) differentiators with fixed-time convergence, unmeasurable velocity signals can be reconstructed to eliminate the adverse effects of measurement noises, decreasing the time delay of feedback signals. Moreover, a soft switching strategy is designed to flexibly adjust the switching weights and intervals of fuzzy structures, maintaining smooth control commands. Further, by introducing a nonsingular terminal sliding manifold, tracking errors can rapidly converge to a small neighborhood around the origins within a finite time, and all closed-loop variables are proved to be bounded through the Lyapunov stability theory. Finally, several groups of experiments are carried out on a self-built PAM-driven parallel robot to verify the effectiveness of the suggested method.

Mechanical systems often face unpredictable surrounding situations in applications, which bring lots of intangible uncertainties into system operations. Further, some robot systems, especially, pneumatic artificial muscle (PAM)-driven robot systems, also have accumulative nonlinearities, such as rate-dependent hysteresis, creep, and periodic/regular time-varying parameters, increasing design difficulties of high-accuracy controllers. This paper develops a long short-term memory neural network (LSTM-NN)-enhanced adaptive controller for PAM-driven parallel robot systems with transient and steady-state performance constraints. Specifically, a continuous-time LSTM-NN structure is introduced to recover unknown lumped dynamics, improving the approximation ability of time-dependent terms with accumulative effects. Moreover, a new two-stage error transformation function is designed to flexibly adjust the desired transient and steady-state performance, facilitating better adaptation to task requirements. To our knowledge, this paper proposes the first solution of utilizing the LSTM-NN-based neuroadaptive method for soft actuator-driven robots to enhance tracking accuracy with transient/steady-state performance improvement. The detailed stability analysis and several groups of experimental results on the self-built platform are provided to verify the feasibility and versatility of the proposed method.

Pneumatic artificial muscles (PAMs), as a kind of soft actuators, can overcome compliance limitations of traditional rigid actuators to improve the adaptability of robots. However, some inherent strong nonlinearities and time-varying properties of PAMs, e.g., complex hysteresis and creep, may lead to a lot of control problems. In addition, PAM-driven systems are also faced with input constraints (e.g., deadzones, saturations, and unidirectional inputs), unknown/unmodeled dynamics and external disturbances, which badly degrade the control performance and even cause accidents. Therefore, this paper proposes anewhysteresis compensation-based immersion and invariance (I&I) adaptive fuzzy control method for PAM-driven humanoid robot manipulators, which can approximate the unknown functions and estimate the unknown parameters. To our knowledge, this is thefirstmethod for PAM-driven systems that compensates for system nonlinearities (not onlycomplex hysteresis,but alsoinput deadzones) by utilizing thepriorinformation in inverse hysteresis models, andsimultaneouslyestimates the unknown functions and parameters by designing a fuzzy update law and a parameter update law based on I&I methodology, respectively, which increases thecontrol frequencyof systems and improves tracking performance during high-speed motions. Finally, we apply the proposed approach on a self-built PAM-driven humanoid robot manipulator to validate its effectiveness and robustness.Note to Practitioners—Faced with the practical requirements of robots that interact closely with humans, improving the adaptability and compliance of robots by utilizing soft actuators, such as PAMs, can satisfy current demands. Moreover, PAMs also have many expective characteristics (e.g., high power density, light material, low costs, clean power, etc.), which makes PAMs occupy an important status in the field of soft robotics. However, unknown parameters/structures, strong nonlinearities, and input constraints, may badly degrade the control performance of PAMs. Based on the above characteristics, this paper proposes anewhysteresis compensation-based adaptive fuzzy controller for PAM-driven humanoid robot manipulators, which realizesaccurate trackingcontrol during high-speed motions by using inverse hysteresis models to compensate for strong nonlinearities in PAMs, and a fuzzy update law and a parameter update law based on I&I methodology are utilized to estimate unknown parameters/structures. In addition, the proposed controller cansimultaneouslycompensate for input deadzones by utilizing the hysteresis information, which improves thecontrol frequencyof the manipulators, and can rapidly suppress tracking errors. Experimental results are provided to validate the effectiveness of the presented method. In the future, more effective compensation methods, such as rate-dependent hysteresis models, will further be carried out to compensate for system nonlinearities.

Pneumatic artificial muscle (PAM) actuators exhibit practical compliance and great payload-to-weight ratios when driving robotic exoskeletons. However, filling with highly compressed gas makes PAMs susceptible to sensor noises, which may degrade the state response and increase control efforts. In addition, most of the existing optimal controllers require linearized operations or complex network calculations. To this end, a supervised learning control method with preassigned-time performance is studied, which achieves satisfactory motion control of the compliant PAM robots. In particular, the utilized dynamic observer with time-varying gains significantly reduces the effect of observation noises, and enhances the state convergence speed by combining with the preassigned-time constraints. Simultaneously, the improved supervised learning algorithm further optimizes input air consumption, which only involves the iterative adjustment of network weights. In contrast to the literature, this article presents a new solution to minimize energy consumption of the compliant PAM robots, while ensuring that the output states converge within the preassigned time, independent of parameter design. Rigorous stability analysis is provided and several experiments validate the tracking efficacy of the proposed method.

Magnetic resonance imaging serves as an essential tool for clinical diagnosis, however, suffers from a long acquisition time. Sparse sampling effectively saves this time but images need to be faithfully reconstructed from undersampled data. Among the existing reconstruction methods, the structured low-rank methods have advantages in robustness to the sampling patterns and lower error. However, the structured low-rank methods use the 2D or higher dimension k-space data to build a huge block Hankel matrix, leading to considerable time and memory consumption. To reduce the size of the Hankel matrix, we proposed to separably construct multiple small Hankel matrices from rows and columns of the k-space and then constrain the low-rankness on these small matrices. This separable model can significantly reduce the computational time but ignores the correlation existed in inter- and intra-row or column, resulting in increased reconstruction error. To improve the reconstructed image without obviously increasing the computation, we further introduced the self-consistency of k-space and virtual coil prior. Besides, the proposed separable model can be extended into other imaging scenarios which hold exponential characteristics in the parameter dimension. The in vivo experimental results demonstrated that the proposed method permits the lowest reconstruction error with a fast reconstruction. The proposed approach requires only 4% of the state-of-the-art STDLR-SPIRiT runtime for parallel imaging reconstruction, and achieves the fastest computational speed in parameter imaging reconstruction.

Hybrid beamforming (HBF) is a potential solution to reduce the baseband hardware cost in massive multiple-input multiple-output (MIMO) systems that has drawn considerable attention recently. In this paper, we consider the uplink of a multiuser massive MIMO system in the presence of blockers. Such blockers, which arise from, for example, users served by other non-cooperative base stations (BSs), can limit the system performance if not handled properly. We propose a HBF scheme that can remove the impact of blockers, while preserving signals from intended users. Specifically, in our two-step receive beamforming scheme, the analog beamformer (ABF) is designed based on channel covariance matrices to minimize the powers of blockers, while the digital beamformer (DBF) deals with inter-user interference. We propose an iterative algorithm that efficiently gives a good sub-optimal solution to the NP-hard problem in the ABF design. Moreover, we consider a more complete BS architecture by incorporating automatic gain control (AGC) and analog-to-digital converter (ADC). As we show in the simulations, for a system with full-precision ADCs, our scheme approaches the sum rate of an ideal fully-digital system. Interestingly, for a system with low-resolution ADCs, our HBF scheme can even outperform a fully-digital system.

Radio-frequency (RF) impairments, which intimately exist in wireless communication systems, can severely limit the performance of multiple-input-multiple-output (MIMO) systems. Although we can resort to compensation schemes to mitigate some of these impairments, a certain amount of residual impairments always persists. In this paper, we consider a training-based point-to-point MIMO system with residual transmit RF impairments (RTRI) using spatial multiplexing transmission. Specifically, we derive a new linear channel estimator for the proposed model, and show that RTRI create an estimation error floor in the high signal-to-noise ratio (SNR) regime. Moreover, we derive closed-form expressions for the signal-to-noise-plus-interference ratio (SINR) distributions, along with analytical expressions for the ergodic achievable rates of zero-forcing, maximum ratio combining, and minimum mean-squared error receivers, respectively. In addition, we optimize the ergodic achievable rates with respect to the training sequence length and demonstrate that finite dimensional systems with RTRI generally require more training at high SNRs than those with ideal hardware. Finally, we extend our analysis to large-scale MIMO configurations, and derive deterministic equivalents of the ergodic achievable rates. It is shown that, by deploying large receive antenna arrays, the extra training requirements due to RTRI can be eliminated. In fact, with a sufficiently large number of receive antennas, systems with RTRI may even need less training than systems with ideal hardware.

Radio-frequency (RF) impairments in the transceiver hardware of communication systems (e.g., phase noise (PN), high power amplifier (HPA) nonlinearities, or in-phase/quadrature-phase (I/Q) imbalance) can severely degrade the performance of traditional multiple-input multiple-output (MIMO) systems. Although calibration algorithms can partially compensate these impairments, the remaining distortion still has substantial impact. Despite this, most prior works have not analyzed this type of distortion. In this paper, we investigate the impact of residual transceiver hardware impairments on the MIMO system performance. In particular, we consider a transceiver impairment model, which has been experimentally validated, and derive analytical ergodic capacity expressions for both exact and high signal-to-noise ratios (SNRs). We demonstrate that the capacity saturates in the high-SNR regime, thereby creating a finite capacity ceiling. We also present a linear approximation for the ergodic capacity in the low-SNR regime, and show that impairments have only a second-order impact on the capacity. Furthermore, we analyze the effect of transceiver impairments on large-scale MIMO systems; interestingly, we prove that if one increases the number of antennas at one side only, the capacity behaves similar to the finite-dimensional case. On the contrary, if the number of antennas on both sides increases with a fixed ratio, the capacity ceiling vanishes; thus, impairments cause only a bounded offset in the capacity compared to the ideal transceiver hardware case.

Radio-frequency (RF) impairments, that exist intimately in wireless communications systems, can severely degrade the performance of traditional multiple-input multiple-output (MIMO) systems. Although compensation schemes can cancel out part of these RF impairments, there still remains a certain amount of impairments. These residual impairments have fundamental impact on the MIMO system performance. However, most of the previous works have neglected this factor. In this paper, a training-based MIMO system with residual transmit RF impairments (RTRI) is considered. In particular, we derive a new channel estimator for the proposed model, and find that RTRI can create an irreducible estimation error floor. Moreover, we show that, in the presence of RTRI, the optimal training sequence length can be larger than the number of transmit antennas, especially in the low and high signal-to-noise ratio (SNR) regimes. An increase in the proposed approximated achievable rate is also observed by adopting the optimal training sequence length. When the training and data symbol powers are required to be equal, we demonstrate that, at high SNRs, systems with RTRI demand more training, whereas at low SNRs, such demands are nearly the same for all practical levels of RTRI.