Artificial muscles (AMs) traditionally rely on pneumatic sources of fluid power. The use of hydraulics can increase the power and force to weight and volume ratios of AM actuators. This paper develops a control-centric third-order single-input single-output (SISO) lumped-parameter dynamic model and sliding mode position controller based on Filippov's principle of equivalent dynamics for a braided hydraulic artificial muscle (HAM) actuator. The model predicts the nonlinear behavior of the HAM free contraction and captures the fluid and actuator nonlinear dynamic interactions in addition to the braid deformation. Model simulations are compared to experimental results for quasi-static pressurization, isometric pressurization, and open-loop square wave commands at 0.25, 0.5, and 1 Hz. Experiments of sine wave tracking at 0.25, 0.5, and 1 Hz and continuous square wave tracking at 0.067 Hz are conducted using a sliding mode controller (SMC) derived from the model. The SMC achieves a steady-state error of 6 μ m at multiple setpoints within the actuator's 17 mm stroke. Compared to a proportional-integral-derivative (PID) controller, the SMC root-mean-square (RMS) error, mean error, and absolute maximum error are reduced on average by 53%, 61%, and 44%, respectively, demonstrating the benefit of model-based approaches for controlling HAMs.

Pneumatic artificial muscles (PAMs) are a compelling actuator for physical human–robot interaction (pHRI) due to their low mass, high-force capacity, and muscle-like characteristics. However, their low efficiency and bandwidth have forced mobile robotics researchers to examine alternative actuators for performing dynamic tasks like walking and grasping. Recently, the sleeve PAM, has been proposed and shown to improve the efficiency and force capacity when compared with traditional PAM designs. However, the increase in the dynamic performance of sleeve PAMs has not yet been studied. The aim of this research is to compare the dynamic performance of sleeve and traditional PAMS, and to develop a phenomenological model of their dynamic performance. Testing found that the isometric bandwidth of sleeve muscles can be 100% greater than that of traditional muscles at rest length if pressure response is considered, although this improvement decreases with contraction. If force is instead considered, the increase in bandwidth is even greater (up to 120% greater than that of a traditional PAM). The volume of both PAMs was determined using an experimental method, and a phenomenological model was fitted. When these models were used to simulate the performance of a PAM-actuated system, it was shown that both approximate the behavior of the measured system with good accuracy. Finally, a proposed implementation is given which illustrates how the benefits of the sleeved PAM actuator design could be realized in a practical robotics application.

For human–machine interaction, the forward progression of technology, particularly controls, regularly brings about new possibilities. Indeed, healthcare applications have flourished in recent years, including robotic rehabilitation, exercise, and prosthetic devices. Testing these devices with human subjects is inherently risky and frequently inconsistent. This work offers a novel simulation framework toward overcoming many of these difficulties. Specifically, generating a closed-loop dynamic model of a human or a human subsystem that can connect to device simulations allows simulated human–machine interaction. In this work, a muscle-actuated open kinematic chain linkage is generated to simulate the human, and a backstepping controller based on inverse dynamics is derived. The control architecture directly addresses muscle redundancy, and two options to resolve this redundancy are evaluated. The specific case of a muscle-actuated arm linkage is developed to illustrate the framework. Trajectory tracking is achieved in simulation. The muscles recruited to meet the tracking goal are in agreement with the method used to solve the redundancy problem. In the future coupling such simulations to any relevant simulation of a machine will provide safe, insightful preprototype test results.

Pneumatic artificial muscles (PAMs) are an important type of bionic actuators for high power-to-weight ratio and good flexibility. However, nonlinearities always exist in systems driven by PAMs, which should be dealt with to obtain good performances during position control. In this paper, a nonlinear state error feedback controller (NSEFC) is presented to obtain good position accuracy for a two-joint manipulator system driven by PAMs. The two-joint manipulator system has nonlinearities, which come from friction, hysteresis, uncertainties of model, loss of piping pressure, coupling between two joints, and so on. A nonlinear extended state observer (ESO) is designed to estimate the nonlinearities in the two-joint manipulator system. Both the effectiveness of the nonlinear ESO and stability of the two-joint manipulator system are given by Lyapunov approaches. Experimental results are obtained to show that the position accuracy of the two-joint manipulator system is improved based on the proposed method in this paper.

The high force-to-weight ratios of braided fluidic artificial muscles (AMs) are ideal for human scale and mobile robot applications. Prior modeling efforts focus on the theoretical static characteristics or empirical dynamic models of these actuators when pressurized. This paper develops a comprehensive high fidelity theoretical dynamic model based on first principles for braided pneumatic AMs and presents experimental validation. A novel theoretical model for the nonlinear stiffness is derived to describe the pressure–displacement behavior. The stiffness model, together with friction, damping, and inertia models, forms an equation of motion (EOM) for braided pneumatic AMs. The EOM is coupled with first-order servopneumatic pressure dynamics, resulting in a third-order system model. System model simulations are compared to experimental results of prototypes with nine different geometries. On average, the system model is able to predict the quasi-static displacement within 7% and the dynamic response within 11%. The theoretical model is also benchmarked against a high fidelity curve fit method, with the empirical method showing a 2% improvement in only quasi-static scenarios. The model promises to be useful for mechanical system and model-based control designs.

Pneumatic artificial muscles (PAMs) are an interesting type of actuators as they provide high power-to-weight and power-to-volume ratio. However, their efficient use requires very accurate control methods taking into account their complex and nonlinear dynamics. This paper considers a two degrees-of-freedom platform whose attitude is determined by three pneumatic muscles controlled by servovalves. An overactuation is present as three muscles are controlled for only two degrees-of-freedom. The contribution of this work is twofold. First, whereas most of the literature approaches the control of systems of similar nature with sliding mode control, we show that the platform can be controlled with the flatness-based approach. This method is a nonlinear open-loop controller. In addition, this approach is model-based, and it can be applied thanks to the accurate models of the muscles, the platform and the servovalves, experimentally developed. In addition to the flatness-based controller, which is mainly a feedforward control, a proportional-integral (PI) controller is added in order to overcome the modeling errors and to improve the control robustness. Second, we solve the overactuation of the platform by an adequate choice for the range of the efforts applied by the muscles. In this paper, we recall the basics of this control technique and then show how it is applied to the proposed experimental platform. At the end of the paper, the proposed approach is compared to the most commonly used control method, and its effectiveness is shown by means of experimental results.

In this paper, we introduce two robust adaptive controllers for the human shank motion tracking problem that is inherent in neuromuscular electrical stimulation (NMES) systems. The control laws adaptively compensate for the unknown parameters that appear nonlinearly in the musculoskeletal dynamics while providing robustness to additive disturbance torques. The adaptive schemes exploit the Lipschitzian and/or the concave/convex parameterizations of the model functions. The resulting control laws are continuous and guarantee practical tracking for the shank angular position. The performance of the two robust adaptive controllers is demonstrated via simulations.

A simulation study to control the motion of a human arm using muscle excitations as inputs is presented to validate a recently developed adaptive output feedback controller for a class of unknown multi-input multi-output (MIMO) systems. The main contribution of this paper is to extend the results of Nguyen and Leonessa (2014, “Adaptive Predictor-Based Output Feedback Control for a Class of Unknown MIMO Linear Systems,” ASME Paper No. DSCC2014-6214; 2014, “Adaptive Predictor-Based Output Feedback Control for a Class of Unknown MIMO Linear Systems: Experimental Results,” ASME Paper No. DSCC2014-6217; and 2015, “Adaptive Predictor-Based Output Feedback Control for a Class of Unknown MIMO Systems: Experimental Results,” American Control Conference, pp. 3515–3521) by combining a recently developed fast adaptation technique and a new controller structure to derive a simple approach for a class of high relative degree uncertain systems. Specifically, the presented control approach relies on three components: a predictor, a reference model, and a controller. The predictor is designed to predict the systems output for any admissible control input. A full state feedback control law is then derived to control the predictor output to approach the reference system. The control law avoids the recursive step-by-step design of backstepping and remains simple regardless of the system relative degree. Ultimately, the control objective of driving the actual system output to track the desired trajectory is achieved by showing that the system output, the predictor output, and the reference system trajectories all converge to each other. Thelen and Millard musculotendon models (Thelen, D. G., 2003, “Adjustment of Muscle Mechanics Model Parameters to Simulate Dynamic Contractions in Older Adults,” ASME J. Biomech. Eng., 125 (1), pp. 70–77; Millard, M, Uchida, T, Seth, A, and Delp, Scott L., 2013, “Flexing Computational Muscle: Modeling and Simulation of Musculotendon Dynamics,” ASME J. Biomech. Eng., 135 (2), p. 021005) are used to validate the proposed controller fast tracking performance and robustness.

Muscle fatigue is a neuromuscular condition experienced during daily activities. This phenomenon is generally characterized using surface electromyography (sEMG) signals and has gained a lot of interest in the fields of clinical rehabilitation, prosthetics control, and sports medicine. sEMG signals are complex, nonstationary and also exhibit self-similarity fractal characteristics. In this work, an attempt has been made to differentiate sEMG signals in nonfatigue and fatigue conditions during dynamic contraction using multifractal analysis. sEMG signals are recorded from biceps brachii muscles of 42 healthy adult volunteers while performing curl exercise. The signals are preprocessed and segmented into nonfatigue and fatigue conditions using the first and last curls, respectively. The multifractal detrended moving average algorithm (MFDMA) is applied to both segments, and multifractal singularity spectrum (SSM) function is derived. Five conventional features are extracted from the singularity spectrum. Twenty-five new features are proposed for analyzing muscle fatigue from the multifractal spectrum. These proposed features are adopted from analysis of sEMG signals and muscle fatigue studies performed in time and frequency domain. These proposed 25 feature sets are compared with conventional five features using feature selection methods such as Wilcoxon rank sum, information gain (IG) and genetic algorithm (GA) techniques. Two classification algorithms, namely, k-nearest neighbor (k-NN) and logistic regression (LR), are explored for differentiating muscle fatigue. The results show that about 60% of the proposed features are statistically highly significant and suitable for muscle fatigue analysis. The results also show that eight proposed features ranked among the top 10 features. The classification accuracy with conventional features in dynamic contraction is 75%. This accuracy improved to 88% with k -NN-GA combination with proposed new feature set. Based on the results, it appears that the multifractal spectrum analysis with new singularity features can be used for clinical evaluation in varied neuromuscular conditions, and the proposed features can also be useful in analyzing other physiological time series.

This work combines the kinematics estimate of human standing with a hybrid identification algorithm to identify a set of ankle dynamics mechanical parameters. We used the hold and release (H&R) experimental paradigm to model a set of recoverable falls on a population of unimpaired adults. Body kinematics was acquired with a microsoft kinect ( mk ) version 2 after benchmarking its position accuracy to a camera-based vision system (CVS). The system identification algorithm, combining an extended Kalman filter (EKF) and a genetic algorithm (GA), allowed to identify the effect of tendon and muscle stiffness at the ankle joint, separately. This work highlights that, when associated to soft-computing techniques, affordable tracking devices developed for the gaming industry can be used for the reliable assessment of neuromechanical parameters in clinical settings.

This paper describes the design and control of a robotic elbow system, which is actuated with a novel sleeve muscle actuator. The sleeve muscle is a significant step forward from the traditional pneumatic muscle, and provides a substantially improved performance through a fundamental structural change. Specifically, the sleeve muscle incorporates a cylindrical insert to the center of the pneumatic muscle, which eliminates the central portion of the internal volume. As a result of this change, the sleeve muscle provides multiple advantages over the traditional pneumatic muscle, including the increased force capacity over the entire range of motion, reduced energy consumption, and expedited dynamic response. Furthermore, utilizing the load-bearing tube as the insert, the sleeve muscle enables an innovative “actuation-load bearing” structure, which generates a highly compact robotic system to mimic the structure and functionality of biological limbs. The robotic elbow design in this paper serves an example that shows the design and control process of a robotic joint in this integrated structure. This robotic elbow provides a range of motion of 110 deg, approximately 80% of that for a human elbow, and an average torque capacity that exceeds the peak torque of the human elbow. The servo control capability is provided with a model-based sliding-mode control approach, which is able to provide good control performance in the presence of disturbances and model uncertainties. This controller is implemented on the robotic elbow prototype, and the effectiveness was demonstrated with step response and sinusoidal tracking experiments.

Pneumatic muscle actuators offer a higher force-to-weight ratio compared to traditional cylinder actuators, and introduce stick-slip-free operation that offers an interesting option for positioning systems. Despite several advantages, pneumatic muscle actuators are commonly avoided in industrial applications, mainly due to rather different working principles. Due to the highly nonlinear characteristics of the muscle actuator and pneumatic system, a reliable control strategy is required. Although muscle actuators are widely studied, the literature lacks detailed studies where the performance for servo systems is compared with traditional pneumatic cylinders. In this paper, a pneumatic servo actuation system is compared with a traditional cylinder actuator. As the overall system dynamics are highly nonlinear and not well defined, a sliding mode control (SMC) strategy is chosen for the control action. In order to improve the tracking performance, an SMC strategy with an integral action (SMCI) is also implemented. The control algorithms are experimentally applied on the pneumatic muscle and the cylinder actuator, for the purposes of position tracking. The robustness of the systems are verified and compared by varying the applied loads.

Pneumatic artificial muscles (PAMs) are comprised of an elastomeric bladder surrounded by a braided mesh sleeve. When the bladder is inflated, the actuator may either contract or extend axially, with the direction of motion dependent on the orientation of the fibers in the braided sleeve. Contractile PAMs have excellent actuation characteristics, including high specific power, specific work, and power density. Unfortunately, extensile PAMs exhibit much reduced blocked force, and are prone to buckling under axial compressive loading. For applications in which extensile motion and compressive force are desired, the push-PAM actuator introduced here exploits the operational characteristics of a contractile PAM, but changes the direction of motion and force by employing a simple internal mechanism using no gears or pulleys. Quasi-static behavior of the push-PAM was compared to a contractile PAM for a range of operating pressures. Based on these data, the push-PAM actuator can achieve force and stroke comparable to a contractile PAM tested under the same conditions.

Stable stochastic feedback control of an aggregate output from a multitude of cellular units is presented in this paper. Similar to a skeletal muscle comprising a number of muscle fibers, the plant considered in this paper consists of many independent units (called cellular units), each of which contributes to an aggregate output of the whole system. The central controller regulates the aggregate output by stochastically recruiting as many cellular units as needed for producing a required output. Two challenges are considered. The first is how to deal with individual units having pronounced hysteresis and long latency time in transient response. It will be shown that slow response and poor stability due to the hysteresis and latency time can significantly be improved by coordinating the multitude of cellular units, which are in diverse phases in the hysteresis loop. The second challenge is how to build a central controller that coordinates the multitude of cellular units without knowing the state of individual units. Stochastic broadcast feedback is presented as a solution that meets those requirements. The central controller observes only the aggregate output value rather than the output and state of each unit, compares the aggregate output against a reference, and broadcasts an error signal to all the units, which are anonymous. In turn, each cellular unit makes a control decision stochastically with state transition probabilities that are modulated by the broadcast error signal from the central controller. Stability analysis based on supermatingale theory guarantees that this stochastic broadcast feedback is stable and robust against cell failures. The method is applied to the control of shape-memory-alloy muscle actuators with cellular architecture. Despite pronounced hysteresis and long latency time, stochastic broadcast feedback can achieve fast and stable control. Simulation experiments verify the theoretical results.

This paper describes the design and control of a new chemomuscle actuation system for robotic systems, especially the mobile systems inspired by biological principles. Developed based on the pneumatic artificial muscle, a chemomuscle actuation system features a high power density, as well as similar characteristics to the biological muscles. Furthermore, by introducing monopropellant (a special type of liquid fuel) as the energy storage media, the chemomuscle system leverages the high energy density of liquid fuel and provides a compact form of high-pressure gas supply with a simple structure. The introduction of monopropellant addresses the limitation of pneumatic supply on mobile devices and thus is expected to facilitate the future application of artificial muscle on biorobotic systems. In this paper, the design of a chemomuscle actuation system is presented, as well as a robust controller design that provides effective control for this highly nonlinear system. To demonstrate the proposed chemomuscle actuation system, an experimental prototype is constructed, on which the proposed control algorithm provides good tracking performance.

A new approach to controlling the ensemble behavior of many identical agents is presented in this paper, inspired by motor recruitment in skeletal muscles. A group of finite state agents responds randomly to broadcast commands, each producing a state-dependent output that is measured in aggregate. Despite the lack of feedback signal and initial state information, this control architecture allows a single central controller to direct the aggregate output of the ensemble toward a desired value. First, the system is modeled as an ensemble of statistically independent, identically distributed, binary-state Markov processes with state transition probabilities designated by a central controller. Second, steady-state behavior, convergence rate, and variance of the aggregate output, i.e., the total number of recruited agents, are analyzed, and design trade-offs in terms of accuracy, convergence speed, and the number of spurious transitions are made. Third, a limited feedback signal, only detecting if the output has reached a goal, is added to the system, and the recruitment controller is designed as a stochastic shortest path problem. Optimal convergence rate and associated transition probabilities are obtained. Finally, the theoretical results are verified and demonstrated with both numerical simulation and control of an artificial muscle actuator made up of 60 binary shape memory alloy motor units.

This paper presents a sliding mode controller for a 2DOF planar pneumatic manipulator actuated by pleated pneumatic artificial muscle actuators. It is argued that it is necessary to account for the pressure dynamics of muscles and valves. A relatively detailed system model that includes pressure dynamics is established. Since the model includes actuator dynamics, feedback linearization was necessary to design a sliding mode controller. The feedback linearization and subsequent controller design are presented in detail, and the controller’s performance is evaluated, both in simulation and experimentally. Chattering was found to be quite severe, so the introduction of significant boundary layers was required.

A mathematical driver model is introduced in order to explain the driver steering behavior observed during successive double lane-change maneuvers. The model consists of a linear quadratic regulator path-following controller coupled to a neuromuscular system (NMS). The NMS generates the steering wheel angle demanded by the path-following controller. The model demonstrates that reflex action and muscle cocontraction improve the steer angle control and thus increase the path-following accuracy. Muscle cocontraction does not have the destabilizing effect of reflex action, but there is an energy cost. A cost function is used to calculate optimum values of cocontraction that are similar to those observed in the experiments. The observed reduction in cocontraction with experience of the vehicle is explained by the driver learning to predict the steering torque feedback. The observed robustness of the path-following control to unexpected changes in steering torque feedback arises from the reflex action and cocontraction stiffness of the NMS. The findings contribute to the understanding of driver-vehicle dynamic interaction. Further work is planned to improve the model; the aim is to enable the optimum design of steering feedback early in the vehicle development process.

This paper is concerned with developing a method to measure a driver’s muscle activation strategy during driving, to enable validation of a mathematical model. Electromyography (EMG) was successfully used to measure driver muscle activity. Regression analysis of the EMG data, measured from the right arm of one test subject, was used to determine the key muscles involved in generating steering torque. The significant muscles were found to be the mid and front deltoid, the sternal portion of the pectoralis major, and the triceps long head. Using the identified regression parameters and measured EMG from key muscles, forces generated by the driver at the rim of the steering wheel were predicted under isometric conditions. The method was further developed and measurements were taken from the left and right arms of eight test subjects. Using regression analysis a model that predicts steering torque from the EMG signals was generated. The method also allows co-contraction of opposing muscles to be identified during a dynamic steering maneuver. Muscle co-contraction is thought to be a significant control strategy employed by drivers, and is the subject of further work.

Programmable mechanical compliance in actuation is desirable for human interaction tasks and important for producing biomimetic motion, particularly for robots designed for use in domestic settings. In this paper, the equilibrium point (EP) hypothesis is proposed and implemented as a new strategy for controlling programmable compliance. The primary objective of this work is to design and demonstrate a simple robot control strategy that can potentially be used by assistive robots to learn and execute compliant interaction tasks from human demonstrations. A 2-DOF planar manipulator activated by McKibben actuators was constructed for the purpose of demonstrating the application of the EP hypothesis on an inexpensive robotic platform, such as might be used in domestic applications. The equilibrium angle and stiffness of each of the joints on the manipulator can be independently programmed. The results presented herein show stable and satisfactory tracking behavior during free motion, interaction, and transition tasks for a robot control system inspired by the EP hypothesis and implemented with a linear proportional-integral (PI) control strategy.