Mahdi Haghshenas-Jaryani

Modeling and sliding mode control of a snake-like robot with holonomic constraints

In this paper modeling and robust sliding mode control of a snake-like robot in tracking of serpenoidal motion is addressed. By considering holonomic constraints, the kinematic and dynamic equations of a hyper redundant robot are obtained using Kanes method. The Coulomb friction between robots wheels and the ground is modeled in two perpendicular directions (i.e. tangential and normal directions) with unknown coefficients of friction in each direction. A sliding manifold is defined by using position and velocity tracking errors of the relative motion between consecutive joints. Continuous approximation of switching control laws is used to eliminate chattering phenomena. The simulation results demonstrate that the snake-like robot realizes the trajectory tracking with desired robustness to the parameter uncertainties.

Multiscale dynamic modeling of processive motor proteins

This paper examines dynamic behavior of a single scale biological system, nanoscale processive motor proteins, which has multiscale features because of interactions with surrounding environment. A new multiscale mechanical modeling approach has been introduced that addresses these interactions, including contact and impact with substrate and fluidic viscosity, during motor proteins processivity. The idea leads to retaining mass properties which are critical to the treatment of contact and impact required for motor protein processivity. In addition, the model supports the intriguing possibility that the effective viscous friction forces are quite small. In order to accomplish this, minimal form of the first and the second order models using embedded constraints are developed. A comparison between a widely-accepted overdamped and the proposed underdamped dynamic behavior is presented. The resulting motion predicted by the spatial model of motor proteins are obtained using a simplified model of Myosin V.

Sensorized soft robotic glove for continuous passive motion therapy

This paper presents the design and development of a sensorized soft robotic glove based on pneumatic soft-and-rigid hybrid actuators for providing continuous passive motion (CPM) in hand rehabilitation. This hybrid actuator is comprised of bellow-type soft actuator sections connected through block-shaped semi-rigid sections to form robotic digits. The actuators were designed to satisfy the anatomical range of motion for each joint. Each digit was sensorized at the tip with an inertial measurement unit sensor in order to track the rotation of the distal end. A pneumatic feedback control system was developed to control the motion of the soft robotic digit in following desired trajectories. The performance of the soft robotic glove and the associated control system were examined on an able-bodied subject during flexion and extension to show the gloves applicability to CPM applications.

Design and Development of a Novel Soft-and-Rigid Hybrid Actuator System for Robotic Applications

This paper presents the design and development of a pneumatic soft-and-rigid hybrid actuator system that consists of half-bellow shaped soft sections in-between block shape rigid sections. The hybrid actuator architecture allows for selective actuation of each soft section (acting as a joint) with precise control over its bending motion. The soft half-bellow section is designed as a series of hollow ridges extending straight to a flat base. This geometry provides forward and backward bending motion when subjected to positive and negative pressure, respectively. Bending occurs as the ridges of the soft section expand and contract more than the flat base due to pressure variations. The rigid sections serve as connections between soft actuator sections and enhance force transfer. As a case study, a hybrid actuator system was designed as a soft robotic digit with three soft joints and four rigid connecting sections. Finite element analysis was performed to evaluate the design parameters such as number of ridges and materials for the robotic finger. The joints (from proximal to distal) were designed to have four, three, and two ridges, respectively, to generate the desired range of angular motion. Fabrication of the finger was done with silicone rubber RTV-4234-T4 and PMC polyurethane rubber using a combination of compression molding and overmolding processes. The angular and translational displacements of the robotic finger were experimentally and numerically evaluated at different pressures. The trajectory of the fingertip is comparable to those reported in literature for continuous soft actuators with a similar length. The significance of this actuator system is that both range of angular and translational motions are achieved at low pressure, less than 70kPa, as opposed to reported pressures of greater than 100kPa. The presented results show the great potential of the soft robotic finger for use in robotic, rehabilitation, and assistive device applications.Copyright

Multiscale dynamic modeling of flexibility in myosin V using a planar mechanical model

This paper presents a modification to the multiscale dynamic approach, introduced in [1]–[5], for the simulation and analysis of flexibility in motor proteins, especially myosin V. In the previous work, a new multiscale dynamic modeling approach has been developed that dissolves the issue of the long simulation run time that is due to the disproportionality between the small mass of myosin V relative to the large viscous drag coefficient. The interesting aspect of the approach is that it retains the mass properties, in contrast to the commonly used models which omit mass properties, at the nanoscale, to address the disproportionality issue. This paper discusses modeling flexibility in the protein as an extension of the original rigid multibody model. Adding flexibility to the mechanical model of motor protein creates an extra disproportionate issue between the mass (0.48ag), the viscous damping coefficient (108 ag/ms), and the stiffness constant (1014 ag.nm2/ms2) which cannot be handled by the original multiscale approach. The proposed modification helps to address the issue by introducing an extra scaling factor that brings all generalized active forces into proportion with the inertial terms. In order to show the effectiveness of the modified approach, a flexible mechanical model of myosin V is developed. Empirical studies have shown that myosin Vs neck domain can be considered as three pairs of tandem elements called IQ motifs which can bend at junctures between them. Therefore, each neck is modeled by three rigid bodies connected by flexible pin joints together, rather than a single rigid body has been used in the previous works [1]–[5].

Trajectory Control of Snake-Like Robots in Operational Space Using a Double Layer Sliding Mode Controller

This paper presents operational space control of snake-like robots for tracking designated paths using a double layer sliding mode control algorithm. The snake robot has n links with assumed lateral sliding that leads to n+2 degrees-of-freedom (DOF) while it has only n-1 actuator at joints and therefore it is underactuated. Kinematic constraints were determined which describe the geometric relationship between the position of links’ mass center and the joints’ relative angle. The outer layer (loop) of the controller was designed to modulate the parameters of the serpenoid curve using the kinematic constraints in order to the mass center of links follow different designated paths. The inner layer of sliding mode controller was developed to guarantee tracking of the modulated serpenoidal pattern by the snake robot’s joints. In this work, Kane’s method was used to model the robot dynamics with a Coulomb friction for interaction with ground. Uncertainty with an upper bound was considered for the model parameters. To demonstrate the effectiveness of the designed controller in presence of uncertainties, the double layer controller was examined on a four links snake-like robot with uncertain model parameters in tracking of a straight line and a circular path. Simulation results are presented in support of the proposed idea.Copyright

Multiscale Dynamic Modeling of Flexibility in Myosin V

This paper presents a multiscale dynamic model for the simulation and analysis of flexibility in myosin V. A three dimensional (3D) flexible multibody model is developed to mechanically model the biological structure of myosin V. Experimental studies have shown that myosin’s neck domain can be considered as three pairs of tandem elements which can bend at junctures between them. Therefore, each neck is modeled by three rigid bodies connected by flexible spherical joints. One of the most important issues in dynamic modeling of micro-nanoscale sized biological structures, likes DNA and motor proteins, is the long simulation run time due to the disproportionality between physical parameters involved in their dynamics such as mass, drag coefficient, and stiffness. In order to address this issue, the mostly used models, based on the famous overdamped Langevin dynamics, omit the inertial terms in the equations of motion; that leads to a first order model which is inconsistent with the Newton’s second law. However, the proposed model uses the concept of the method of multiple scales (MMS) that brings all terms of the equations of motion into proportion with each other that helps to retain the inertia terms. This keeps consistency of the model with the physical laws and increases time step size of numerical integration from commonly used sub-femto seconds to sub-milli seconds. Therefore, simulation run time will be many orders of magnitude less than ones based on the other approaches. The simulation results obtained by the proposed multiscale model show more realistic dynamic behavior of myosin V in compared with other models.© 2013 ASME

Multiscale modeling and simulation of a microbead in an optical trapping process

The purpose of this work is to generate a theoretical model for the dynamics of a polystyrene microsphere under the influence of Gaussian beam optical tweezers (OTs) in the ray-optics regime. OTs use the radiation pressure from a focused laser beam to manipulate microscopic objects as small as atoms [1]. They have been used in the biological sciences to measure nanometer-range displacements, apply picoNewton-range forces, and determine the mechanical properties of DNA, cell membranes, whole cells, and microtubules. The proposed model takes into account the forces and moments imparted onto the microbead by the OTs beam, and uses a Newton-Euler Dynamics framework to generate the equations of motion. Although examination of dimensionless numbers and other indicators including, Reynolds number 10−9 ≤ Re ≤ 10−4, Knudsen number 0.0001875, and the disproportionality between the mass and the viscous drag co-efficients O(10−4), does not clearly indicate whether this is a multiscale problem or not; but, a numerical integration of the original model leads to a long simulation run-time, a few days. Moreover, investigation of the step size showed that the adaptive numerical integrator was proceeding with a picosecond step size in order to achieve the requested accuracy. This situation implies a multiscale feature involved in the dynamics of optical trapping process of the small bead. To address this issue, a multiscale model is developed that helps to significantly reduce the simulation run-time and reveals underdamped behavior of the bead. In order to verify the theoretical model, experiments were carried out on a microsphere bead with 1.6μm diameter. A comparison of experimental data and simulation data indicate that this approach closely models microparticle behavior to the accuracy of the experiment under Gaussian beam optical tweezers.Copyright

A Supervisory Fuzzy-PID Controller for a MIMO Biped Robot Balance in Frontal Plane

In this paper we propose to control a bipedal robot in an unstable position by means of a PID controller that gains are turned by a fuzzy logic system. For that, a model of planar 3 linked segment consisting of limb, trunk and extended arms with fixed base is used. Fuzzy if-then rules are constructed based on human expert knowledge and biomechanics studies for tuning of PID’s gain. For construction of tuning rules, we have developed an optical measuring system to record experimental data of balance keeping of a human in an unstable position. The control model is based on three sets of different global variables: (1) limb orientation and its derivative, (2) trunk/upper attitude and its derivative, and (3) orientation of extended arms and its derivative. In this study, we consider only side to side robot’s balancing and we also demonstrate, by simulation results, that the proposed control model is stable in presence of external disturbance. Finally we have compared the simulation results with the experiments to show the similarity of the proposed system with the human balance keeping task.Copyright