abstract: Robotic joints can be either powered or passive. This work will discuss the creation of a passive and a powered joint system as well as the combination system being both powered and passive along with its benefits. A novel approach of analysis and control of the combination system is presented. A passive and a powered ankle joint system is developed and fit to the field of prosthetics, specifically ankle joint replacement for able bodied gait. The general 1 DOF robotic joint designs are examined and the results from testing are discussed. Achievements in this area include the able bodied gait like behavior of passive systems for slow walking speeds. For higher walking speeds the powered ankle system is capable of adding the necessary energy to propel the user forward and remain similar to able bodied gait, effectively replacing the calf muscle. While running has not fully been achieved through past powered ankle devices the full power necessary is reached in this work for running and sprinting while achieving 4x’s power amplification through the powered ankle mechanism. A theoretical approach to robotic joints is then analyzed in order to combine the advantages of both passive and powered systems. Energy methods are shown to provide a correct behavioral analysis of any robotic joint system. Manipulation of the energy curves and mechanism coupler curves allows real time joint behavioral adjustment. Such a powered joint can be adjusted to passively achieve desired behavior for different speeds and environmental needs. The effects on joint moment and stiffness from adjusting one type of mechanism is presented. / Dissertation/Thesis / Doctoral Dissertation Mechanical Engineering 2017

Wood pallets are heavily used throughout the United States and the World to transport, store, and protect goods. During a lifecycle, pallets typically experience various stresses from warehouse storage racks, materials handling equipment, and floor stacking situations. The components within the pallet interact to withstand load and impact forces. Every year product damage and human injury/death result from improperly designed pallets, non-reliable packaging systems, and careless materials handling methods. In use wood pallets are exposed to a variety of loads and support conditions. This research investigates the effect of different pallet designs and support conditions on pallet stiffness. Uniform loads were applied to pallet designs containing thick or thin components and three, four, or five non-notched and notched stringers. The pallets were supported using racked across the length, racked across the width, fork truck tine, and floor stack support conditions. Structural analysis was used to determine the test loads for each pallet bending test. Pallet deflections were measured in specific locations for each bending test. Pallet test results indicated that heavy duty pallets are 6.5 times stiffer than light duty pallets tested in the racked across width (RAW) support condition. Non-notched pallets tested are 51% stiffer than notched pallets in the racked across length (RAL) support condition. Test results also indicated that a wider fork tine support span decreases average pallet stiffness by 29% and 49% for 4 and 5 stringer pallets, compared to 3 stringer. The heavy duty pallets tested are, on average, 48.3% stiffer than light duty pallets in the fork tine support condition. For the notched fork tine support condition, the average pallet stiffness decreased by 29% and 3% for four and five stringer pallets, compared to three stringer. Pallet joints were tested to measure joint stiffness. Joint rotation tests were conducted to determine rotation modulus and joint withdrawal tests were conducted to determine joint withdrawal stiffness. The joint stiffness measurements were used as spring constants in structural analysis based on semi-rigid joint performance. Heavy duty pallet joints were approximately half as stiff (6758 in-lbs/radian) in rotation as light duty pallet joints (12907 in-lbs/radian). Light duty pallet joints were less stiff (44008 lbs/in) in withdrawal than heavy duty pallet joints (57823 in/lbs). The results from this research were used to compare with results from ANSYS (Version 11) structural model estimates. The average predicted error for all pallet bending tests was 13% (heavy duty) and 3% (light duty). / Master of Science

The human foot contributes significantly to the function of the whole lower extremity during standing and locomotion. Nevertheless, the foot and ankle often suffer injuries and are affected by many musculoskeletal and neurological pathologies. The overall aim of this thesis was to evaluate gait parameters and muscle function change due to foot and ankle injury, malalignment and co-contraction. Using 3D gait analysis, analytical analyses and computational simulations, biomechanical consequences of gait impairment at the ankle and foot were explored in ablebodied persons and in patient groups with disorders affecting walking. We have characterized gait patterns of subjects with ankle fractures with a modified multi-segment foot model. The inter-segmental foot kinematics were determined during gait in 18 subjects one year after surgically-treated ankle fractures. Gait data were compared to an age- and gender-matched control group and the correlations between functional ankle score and gait parameters were determined. It was observed that even with fairly good clinical results, restricted range of motion and malalignment at and around the injured area were found in the injured limb. Moment-angle relationship (dynamic joint stiffness) - the relationship between changes in joint moment and changes in joint angle - is useful for demonstrating interaction of kinematics and kinetics during gait. Ankle dynamic joint stiffness during the stance phase of gait was analyzed and decomposed into three components in thirty able-bodied children, eight children with juvenile idiopathic arthritis and eight children with idiopathic toe-walking. Compared to controls, the component associated with changes of ground reaction moment was the source of highest deviation in both pathological groups. Specifically, ankle dynamic joint stiffness differences can be further identified via two subcomponents of this component which are based on magnitudes and rates of change of the ground reaction force and of its moment arm. And differences between the two patient groups and controls were most evident and interpretable here. Computational simulations using 3D musculoskeltal models can be powerful in investigating movement mechanisms, which are not otherwise possible or ethical to measure experimentally. We have quantified the effect of subtalar malalignment on the potential dynamic function of the main ankle dorsiflexors and plantarflexors: the gastrocnemius, soleus and tibialis anterior. Induced acceleration analysis was used to compute muscle-induced joint angular and body center of mass accelerations. A three-dimensional subject-specific linkage model was configured by gait data and driven by 1 Newton of individual muscle force. The excessive subtalar inversion or eversion was modified by offsetting up to ±20˚ from the normal subtalar angle while other configurations remain unaltered. We confirmed that in normal gait, muscles generally acted as their anatomical definitions, and that muscles can create motion in many joints, even those not spanned by the muscles. Excessive subtalar eversion was found to enlarge the plantarflexors’ and tibialis anterior’s function. In order to ascertain the reliability of muscle function computed from simulations, we have also performed a parametric study on eight healthy adults to evaluate how sensitive the muscle-induced joints’ accelerations are to the parameters of rigid foot-ground contact model. We quantified accelerations induced by the gastrocnemius, soleus and tibialis anterior on the lower limb joints. Two types of models, a ‘fixed joint’ model with three fixed joints under the foot and a ‘moving joint’ model with one joint located along the moving center of pressure were evaluated. The influences of different foot-ground contact joint constraints and locations of center of pressure were also investigated. Our findings indicate that both joint locations and prescribed degrees-of-freedom of models affect the predicted potential muscle function, wherein the joint locations are most influential. The pronounced influences can be observed in the non-sagittal plane. Excessive muscle co-contraction is a cause of inefficient or abnormal movement in some neuromuscular pathologies. We have identified the necessary compensation strategies to overcome excessive antagonistic muscle cocontraction at the ankle joint and retain a normal walking pattern. Muscle-actuated simulation of normal walking and induced acceleration analysis were performed to quantify compensatory mechanisms of the primary ankle and knee muscles in the presence of normal, medium and high levels of co-contraction of two antagonistic pairs (gastrocnemiustibialis anterior and soleus-tibialis anterior). The study showed that if the co-contraction level increases, the nearby synergistic muscles can contribute most to compensation in the gastrocnemius-tibialis anterior pair. In contrast, with the soleus-tibialis anterior co-contraction, the sartorius and hamstrings can provide important compensatory roles in knee accelerations. This dissertation documented a broad range of gait mechanisms and muscle functions in the foot and ankle area employing both experiments and computational simulations. The strategies and mechanisms in which altered gait and muscles activation are used to compensate for impairment can be regarded as references for evaluation of future patients and for dynamic muscle functions during gait. / QC 20120514

The field of autonomous walking robots has been dominated by the trajectory-control approach, which rigidly dictates joint angle trajectories at the expense of both energy efficiency and stability, and the passive dynamics approach, which uses no actuators, relying instead on natural mechanical dynamics as the sole source of control. Although the passive dynamics approach is energy efficient, it lacks the ability to modify gait or adapt to disturbances. Recently, minimally actuated walkers, or dynamic walkers, have been developed that use hip or ankle actuators---knees are always passive---to regulate mechanical energy variations through the timely application of joint torque pulses. Despite the improvement minimal actuation has provided, energy efficiency remains below target values and perturbation rejection capability (i.e., stability) remains poor. In this dissertation, we develop and analyze a simplified robotic system to assess biologically inspired methods of improving energy efficiency and stability in dynamic walkers. Our system consists of a planar, dynamically swinging leg with hip and knee actuation. Neurally inspired, nonlinear oscillators provide closed-loop control without overriding the leg's natural dynamics. We first model the passive stiffness of muscles by applying stiffness components to the joints of a hip-actuated swinging leg. We then assess the effect active knee control has on unperturbed and perturbed leg swinging. Our results indicate that passive joint stiffness improves energy efficiency by reducing the actuator work required to counter gravitational torque and by promoting kinetic energy transfer between the shank and thigh. We also found that active knee control 1) is detrimental to unperturbed leg swinging because it negatively affects energy efficiency while producing minimal performance improvement and 2) is beneficial during perturbed swinging because the perturbation rejection improvement outweighs the reduction in energy efficiency. By analyzing the effects of applying passive joint stiffness and active knee control to dynamic walkers, this work helps to bridge the gap between the performance capability of trajectory-control robots and the energy-efficiency of passive dynamic robots.

Concrete-filled tubular (CFT) columns have been widely adopted in building structures owing to their superior structural performance, such as enhanced load bearing capacity, compared to hollow tubes. Circular, square and rectangular hollow sections are most commonly used in the past few decades. Elliptical hollow section (EHS) available recently is regarded as a new cross-section for the CFT columns due to its attractive appearance, optional orientation either on major axis or minor axis and improved structural efficiency. The state of the research in terms of elliptical columns, tubular joints between EHSs and connections with CFT columns, etc., are reviewed in this thesis, showing a lack of investigations on EHSs, especially on beam to elliptical column connections which are essential in framed structures. The structural behaviour of elliptical column to I-beam connections under bending is studied in this thesis to fill the research gap. Overall ten specimens with various joint assemblies were tested to failure to highlight the benefits of adopting concrete infill and stiffeners in the columns. A three-dimensional finite element model developed by using ABAQUS software is presented and verified against obtained experimental results, which shows acceptable accuracy and reliability in predicting failure modes of the connections and their moment capacities. Parametric studies were performed to access the main parameters that affecting the bending behaviour of the connections. A simple hand calculation method in terms of ultimate moment capacity is proposed according to experiments conducted for connections with concrete-filled columns.

Block stability is common and has to be studied in detail for designing tunnels. Stability of block depends upon the shape and size of the blocks, stresses around the block and factors such as clamping forces and the ratio between joint stiffness. These factors are studied in detail and are the main objective of this thesis. In this thesis influence of loading and unloading of blocks on joint stiffness and thus on ultimate pullout force are analyzed. Normal stress on the joint plane is linked with shear stiffness of the joint and relaxation of forces. Changes of forces were considered to estimate joint stiffness and ultimate pullout force using new methods in the present thesis. First method takes into account changing clamping forces considering stiffness ratio constant (Crawford and Bray). The second method was developed in which the ratio between normal and shear stiffness was taken as a function of normal stress (Bagheri and Stille). In third method, gradually pullout force is increased which changes the normal stress and joint stiffness. The lower limit of joint stiffness gives a very conservative design. So a stiffness value based on the average of lower and upper limit of normal force has also been considered. A comparison between the new methods and the previous method proposed by Crawford and Bray which considers a constant ratio of normal and shear stiffness and constant clamping forces shows that Crawford and Bray’s solution overestimates the pullout forces hence the design is unsafe. It was observed that stiffness ratio is an important factor for estimating required rock support and safety.

Falls are the leading cause of injuries among older adults (>65) and frequently result in reduced mobility, loss of independence, decreased quality of life, injury, and death.  Extensive research has been conducted regarding postural coordination and control, and other mechanisms/processes involved in maintaining postural stability.  However, there is relatively limited knowledge regarding the patterns of joint coordination, the underlying postural controller, and efficient methods to assess passive and active musculoskeletal properties relevant to balance.  In the current work, three new methods were developed to address these limitations and also to better understand the effects of localized ankle muscle fatigue, gender, and aging on postural coordination and control. First, two methods were used to evaluate postural coordination.  A wavelet coherence approach was developed and applied to assess the level and pattern of coordination between pairs of joints (i.e., ankle-knee, ankle-trunk, and ankle-head).  In addition, the uncontrolled manifold method was implemented for evaluation of potential whole-body coordination control goals.  Clear patterns of intermittent wavelet coherence were evident, indicating that joint coordination is intermittently executed.  Both in-phase and anti-phase coherence were detected over frequencies of 2.5 -- 4.0 Hz.  Shoulder and head kinematics appeared more likely than the whole-body center of mass as control goals for whole body coordination.  Both aging and ankle muscle fatigue led to a reduction of joint coordination. Second, an intermittent sliding mode controller was developed to model quiet upright stance.  In contrast to most previous postural controllers, which assume continuous control, an intermittent controller was considered more consistent with recent evidence on muscle activity and the results of the first study on postural coordination.  The sliding mode controller was able to accurately track kinematics and kinetics, and generated passive and active ankle torques comparable with previous results.  Ankle fatigue led to an increase in active ankle torque especially among young adults and males. Third, a new method was developed to estimate passive and active mechanical properties at the ankle (e.g., stiffness and damping).  This method was inspired from intermittent control theory, and the earlier results noted.  As opposed to conventional methods, this new method is computationally efficient and does not require external mechanical or sensory perturbations.  The method yielded a ratio of passive to active ankle torques consistent with earlier evidence, and larger passive and active ankle torques among males and older adults.  A post-fatigue increase of active ankle torque was estimated, especially among males and young adults. In addition to providing new analytical methods, the noted studies suggest that older adults have decreased joint coordination and increased ankle stiffness.  As a practical implication of this, fall prevention training programs may benefit from seeking to develop appropriate joint coordination strategies and ankle stiffness magnitudes.  To expand on the current work, future research should consider measuring muscle contraction characteristics at multiple joints and in different postures or activities. / Ph. D.

Concrete-filled tubular (CFT) columns have been widely adopted in building structures owing to their superior structural performance, such as enhanced load bearing capacity, compared to hollow tubes. Circular, square and rectangular hollow sections are most commonly used in the past few decades. Elliptical hollow section (EHS) available recently is regarded as a new cross-section for the CFT columns due to its attractive appearance, optional orientation either on major axis or minor axis and improved structural efficiency. The state of the research in terms of elliptical columns, tubular joints between EHSs and connections with CFT columns, etc., are reviewed in this thesis, showing a lack of investigations on EHSs, especially on beam to elliptical column connections which are essential in framed structures. The structural behaviour of elliptical column to I-beam connections under bending is studied in this thesis to fill the research gap. Overall ten specimens with various joint assemblies were tested to failure to highlight the benefits of adopting concrete infill and stiffeners in the columns. A three-dimensional finite element model developed by using ABAQUS software is presented and verified against obtained experimental results, which shows acceptable accuracy and reliability in predicting failure modes of the connections and their moment capacities. Parametric studies were performed to access the main parameters that affecting the bending behaviour of the connections. A simple hand calculation method in terms of ultimate moment capacity is proposed according to experiments conducted for connections with concrete-filled columns.

The benefits of lighter auto bodies are discussed, and aluminum is compared to steel as an alternative material for auto body construction. The concept of a structural index, lambda, is developed using the simple example of a hollow beam of wall thickness, t, with a cantilever load case. It is shown that the bending stiffness, K, of the beam can be defined as a function of t^lambda, that 1 < lambda < 3, and that lambda can be used to predict the weight savings from material substitution where stiffness is held constant. It is then demonstrated that lambda can be used to predict the weight savings from material substitution in the more complex cases of the joints of a light truck cab.

Master of Science / Pallets are used all over the world in the field of distribution. The strength values associated with a pallet have been thoroughly investigated by many different researchers; however, the stiffness values associated with pallet joints have not. The goal of this work was to investigate the stiffnesses associated with pallets joints made with lag screws and carriage bolts. It is important to understand that different materials, fastening methods, and design considerations can have a huge impact on the stiffness of the joint. This paper will discuss the various tests that were used to measure the actual stiffness of pallet joints and the results of those tests. Afterwards, the researchers detail their attempt to predict the stiffness using an equation created from the actual test data. Finally, by understanding the effects of these various factors, better pallet designs can be created that are both safer and stronger using the investigated alternative fasteners.