Boyin Ding

Real-time FPGA control of a hexapod robot for 6-DOF biomechanical testing

A hexapod robotic test system has been developed to enable complex six degree of freedom (6-DOF) testing of bones, joints, soft tissues, artificial joints and other medical and surgical devices. The device employs six permanent-magnet servomotor driven ballscrews to actuate the system, and measures the displacement response using incremental encoders and loads using a six axis load-cell. The mechanism incorporates a unique design which mitigates many of the issues arising from load-cell compliance, common to most other serial and parallel mechanisms for material testing. This was achieved through a non-collocated design which raises additional challenges. Achieving high bandwidth control of the hexapod also presents challenges, and was achieved using a combination of LabVIEW real-time running on a floating-point Intel processor, along with LabVIEW FPGA running on 16bit Xilinx FPGAs. In this paper the following unique aspects of this hexapod are discussed: the mitigation of load-cell compliance, non-collocated control, implementation of the controller on a real-time platform, and finally technical solutions to solve the complex forward-kinematics solution in real-time. Finally, the results from testing a high-density polymer cylindrical specimen are presented.

Adaptive velocity-based six degree of freedom load control for real-time unconstrained biomechanical testing

Robotic biomechanics is a powerful tool for further developing our understanding of biological joints, tissues and their repair. Both velocity-based and hybrid force control methods have been applied to biomechanics but the complex and non-linear properties of joints have limited these to slow or stepwise loading, which may not capture the real-time behaviour of joints. This paper presents a novel force control scheme combining stiffness and velocity based methods aimed at achieving six degree of freedom unconstrained force control at physiological loading rates.

Stiffness Analysis and Control of a Stewart Platform-Based Manipulator With Decoupled Sensor–Actuator Locations for Ultrahigh Accuracy Positioning Under Large External Loads

Robot frame compliance has a large negative effect on the global accuracy of the system when large external forces/torques are exerted. This phenomenon is particularly problematic in applications where the robot is required to achieve ultrahigh (micron level) accuracy under very large external loads, e.g., in biomechanical testing and high precision machining. To ensure the positioning accuracy of the robot in these applications, the authors proposed a novel Stewart platform-based manipulator with decoupled sensor–actuator locations. The unique mechanism has the sensor locations fully decoupled from the actuator locations for the purpose of passively compensating for the load frame compliance, as a result improving the effective stiffness of the manipulator in six degrees of freedom (6DOF). In this paper, the stiffness of the proposed manipulator is quantified via a simplified method, which combines both an analytical model (robot kinematics error model) and a numerical model [finite element analysis (FEA) model] in the analysis. This method can be used to design systems with specific stiffness requirements. In the control aspect, the noncollocated positions of the sensors and actuators lead to a suboptimal control structure, which is addressed in the paper using a simple Jacobian-based decoupling method under both kinematics- and dynamics-based control. Simulation results demonstrate that the proposed manipulator configuration has an effective stiffness that is increased by a factor of greater than 15 compared to a general design. Experimental results show that the Jacobian-based decoupling method effectively increases the dynamic tracking performance of the manipulator by 25% on average over a conventional method.

Fast and Effective Optimisation of Arrays of Submerged Wave Energy Converters

Renewable forms of energy are becoming increasingly important to consider, as the global energy demand continues to grow. Wave energy is one of these widely available forms, but it is largely unexploited. A common design for a wave energy converter is called a point absorber or buoy. The buoy typically floats on the surface or just below the surface of the water, and captures energy from the movement of the waves. It can use the motion of the waves to drive a pump to generate electricity and to create potable water. Since a single buoy can only capture a limited amount of energy, large-scale wave energy production necessitates the deployment of buoys in large numbers called arrays. However, the efficiency of arrays of buoys is affected by highly complex intra-buoy interactions. The contributions of this article are two-fold. First, we present an approximation of the buoy interactions model that results in a 350-fold computational speed-up to enable the use inside of iterative optimisation algorithms, Second, we study arrays of fully submerged three-tether buoys, with and without shared mooring points.

Effect of degeneration on the six degree of freedom mechanical properties of human lumbar spine segments

While the effects of disc degeneration on compression and rotation motions have been studied, there is no data for shear loading. Clinical research has shown that those with low back pain (a potential consequence of degeneration) experience a 75% greater lateral shear force than those without it. Therefore, the aim was to compare the effect of degeneration on spine segment stiffness and phase angle in each of six degree of freedom (6DOF) loading directions. Fourteen intact functional spinal units (FSU) were dissected from human lumbar spines (mean (SD) age 76.2 (11) years, Thompson grades 3 (N = 5, mild), 4 (N = 6, moderate), 5 (N = 3, severe)). Each FSU was tested in ±6DOFs while subjected to a physiological preload, hydration, and temperature (37°C) conditions in a hexapod robot. A one‐way ANOVA between degenerated groups was performed on stiffness and phase angle for each DOF. Significant differences in stiffness were found between mild and moderate degenerative groups in lateral shear (p = 0.001), and axial rotation (p = 0.001), where moderate degeneration had decreased stiffness. For phase angle, significant differences were seen in anterior shear (p = 0.017), and axial rotation (p = 0.026), where phase angle for mild degeneration was less than moderate. Trends of stiffness and phase angle changes between degenerative groups were similar within each DOF. Clinically, the identification of the DOFs that are most affected by degeneration could be used in rehabilitation to improve supplemental stabilization of core muscle groups.

A novel method to replicate the kinematics of the carpus using a six degree-of-freedom robot

Understanding the kinematics of the carpus is essential to the understanding and treatment of wrist pathologies. However, many of the previous techniques presented are limited by non-functional motion or the interpolation of points from static images at different postures. We present a method that has the capability of replicating the kinematics of the wrist during activities of daily living using a unique mechanical testing system. To quantify the kinematics of the carpal bones, we used bone pin-mounted markers and optical motion capture methods. In this paper, we present a hammering motion as an example of an activity of daily living. However, the method can be applied to a wide variety of movements. Our method showed good accuracy (1.0-2.6°) of in vivo movement reproduction in our ex vivo model. Most carpal motion during wrist flexion-extension occurs at the radiocarpal level while in ulnar deviation the motion is more equally shared between radiocarpal and midcarpal joints, and in radial deviation the motion happens mainly at the midcarpal joint. For all rotations, there was more rotation of the midcarpal row relative to the lunate than relative to the scaphoid or triquetrum. For the functional motion studied (hammering), there was more midcarpal motion in wrist extension compared to pure wrist extension while radioulnar deviation patterns were similar to those observed in pure wrist radioulnar deviation. Finally, it was found that for the amplitudes studied the amount of carpal rotations was proportional to global wrist rotations.

Effect of Potting Technique on the Measurement of Six Degree-of-Freedom Viscoelastic Properties of Human Lumbar Spine Segments

Polymethyl methacrylate (PMMA) and Woods Metal are fixation media for biomechanical testing; however, the effect of each potting medium on the measured six degree-of-freedom (DOF) mechanical properties of human lumbar intervertebral discs is unknown. The first aim of this study was to compare the measured 6DOF elastic and viscoelastic properties of the disc when embedded in PMMA compared to repotting in Woods Metal. The second aim was to compare the surface temperature of the disc when potted with PMMA and Woods Metal. Six human lumbar functional spinal units (FSUs) were first potted in PMMA, and subjected to overnight preload in a saline bath at 37 °C followed by five haversine loading cycles at 0.1 Hz in each of 6DOF loading directions (compression, left/right lateral bending, flexion, extension, left/right axial rotation, anterior/posterior, and lateral shear). Each specimen was then repotted in Woods Metal and subjected to a 2-h re-equilibrating preload followed by repeating the same 6DOF tests. Outcome measures of stiffness and phase angle were calculated from the final loading cycle in each DOF and were expressed as normalized percentages relative to PMMA (100%). Disc surface temperatures (anterior, left/right lateral) were measured during potting. Paired t-tests (with alpha adjusted for multiple DOF) were conducted to compare the differences in each outcome parameter between PMMA and Woods Metal. No significant differences in stiffness or phase angle were found between PMMA and Woods Metal. On average, the largest trending differences were found in the shear DOFs for both stiffness (approximately 35% greater for Woods Metal compared to PMMA) and phase angle (approximately 15% greater for Woods Metal). A significant difference in disc temperature was found at the anterior surface after potting with Woods Metal compared to PMMA, which did not exceed 26 °C. Woods Metal is linear elastic, stiffer than PMMA and may reduce measurement artifact of potting medium, particularly in the shear directions. Furthermore, it is easier to remove than PMMA, reuseable, and cost effective.

Fast and Effective Multi-objective Optimisation of Submerged Wave Energy Converters

Despite its considerable potential, wave energy has not yet reached full commercial development. Currently, dozens of wave energy projects are exploring a variety of techniques to produce wave energy efficiently. A common design for a wave energy converter is called a buoy. A buoy typically floats on the surface or just below the surface of the water, and captures energy from the movement of the waves.

The effect of six degree of freedom loading sequence on the in-vitro compressive properties of human lumbar spine segments

The complex, direction-dependent, poro-viscoelastic properties of the intervertebral disc (disc) suggest that investigations of the six degree of freedom (6DOF) behaviour may be susceptible to inter-test variation in mechanical response if the disc does not return to initial conditions between loading directions. No studies have quantified the effects of sequential multi-directional loading on the consistency of the compressive response of the disc throughout a 6DOF testing protocol. Therefore, the objective of this study was to determine the effect of 6DOF loading on the compressive properties (stiffness and phase angle) of human discs, as evaluated by a reference compression test performed after each single DOF test. Fourteen intact human functional spinal units (FSU) were tested in each of ±6DOFs (shear directions followed by bending and compression) across four orders of magnitude loading frequencies (0.001-1Hz), followed by reference compression tests while subjected to physiological preload, hydration, and body temperature conditions in a hexapod robot. Repeated measures ANOVA revealed significant within-subjects effects between the reference compression tests for modulus (p<0.001), stiffness (p<0.001), and phase angle (p=0.008). Significant post-hoc pairwise comparisons were initially seen between the control and other reference compression tests for stiffness and modulus after the shear DOFs, however, no significant differences were present after the final reference compression test compared to control. More pronounced effects were seen for stiffness in comparison to modulus and phase angle. These effects may be due to three potentials factors, which include the sequence of testing, the cohort of degenerative specimens, and/or cumulative creep due to the constant application of a follower load. While the sequence of test directions was chosen to minimise the biphasic effect, there may be other sequences, which could result in minimal changes in compressive properties.

Assessment of the initial viscoelastic properties of a critical segmental long bone defect reconstructed with impaction bone grafting and intramedullary nailing

INTRODUCTION This study compared the initial viscoelastic properties of a segmental tibial defect stabilized with intramedullary nailing and impaction bone grafting to that of a transverse fracture stabilized with intramedullary nailing. MATERIALS AND METHODS Seven sheep tibiae were tested in compression (1000N), bending and torsion (6Nm) in a six degree-of-freedom hexapod robot. Tests were repeated across three groups: intact tibia (Intact), transverse fracture stabilized by intramedullary nailing (Fracture), and segmental defect stabilized with a nail and impaction bone grafting (Defect). Repeated measures ANOVA on the effect of group on stiffness/phase angle were conducted for each loading direction. RESULTS The Intact group was significantly stiffer than the Fracture and Defect groups in bending and torsion (p<0.022 for both loading directions), and was marginal for the Defect group in compression (p=0.052). No significant differences were found between the Fracture and Defect groups (p>0.246 for all loading directions) for stiffness/phase angle. In compression and bending, phase angles were significantly greater for the Fracture and Defect groups compared to Intact (p<0.025), with no significant differences between groups in torsion (p=0.13). Sensitivity analyses conducted between the Fracture and Defect group differences found that they were not of clinical significance. CONCLUSION The initial properties of a segmental defect stabilized with intramedullary nailing and impaction bone grafting was not clinically significantly different to that of a transverse fracture stabilized with intramedullary nailing.