No description available.

No description available.

Humans need to rapidly select movements that achieve their goal while avoiding negative outcomes. The processes leading to these decisions have only recently been studied. In the typical paradigm used to gain insight into the decision process, participants aim to a target circle that is overlapped by a penalty circle. They receive 100 points for hitting the target, and lose points for hitting the penalty region. Previous research has shown that participants generally behave like a rational decision maker by adapting their endpoint when the distance between the target and penalty circle and the penalty value changes (although some suboptimal selection has been noted). The overall purpose of the research reported in the present thesis was to determine if there are contexts when participants’ behaviour is suboptimal in a rapid, motor decision making tasks. Taken together, the results from four studies showed that: 1) participants require experience and feedback to aim to optimal locations; 2) participants often aimed closer to target center than optimal; and, 3) probability (represented through spatial parameters) has more influence over participant’s motor decisions than does the value of the penalty. Therefore, participants’ actions do not necessarily conform to a rational model of decision making; rather, there are consistent biases arising in the selection, planning, and execution of actions in specific contexts. These findings and conclusions can lead to a more descriptive understanding of motor decision making to provide information that is in addition to prescriptive models of rational behaviour.

Humans need to rapidly select movements that achieve their goal while avoiding negative outcomes. The processes leading to these decisions have only recently been studied. In the typical paradigm used to gain insight into the decision process, participants aim to a target circle that is overlapped by a penalty circle. They receive 100 points for hitting the target, and lose points for hitting the penalty region. Previous research has shown that participants generally behave like a rational decision maker by adapting their endpoint when the distance between the target and penalty circle and the penalty value changes (although some suboptimal selection has been noted). The overall purpose of the research reported in the present thesis was to determine if there are contexts when participants’ behaviour is suboptimal in a rapid, motor decision making tasks. Taken together, the results from four studies showed that: 1) participants require experience and feedback to aim to optimal locations; 2) participants often aimed closer to target center than optimal; and, 3) probability (represented through spatial parameters) has more influence over participant’s motor decisions than does the value of the penalty. Therefore, participants’ actions do not necessarily conform to a rational model of decision making; rather, there are consistent biases arising in the selection, planning, and execution of actions in specific contexts. These findings and conclusions can lead to a more descriptive understanding of motor decision making to provide information that is in addition to prescriptive models of rational behaviour.

Previous research has demonstrated that normally developing children are expected to have adult like control of their speech production skills by 10 years of age, as evidenced by increases in speed of production and decreases in performance variability. (e. g. Kent and Fortier 1980). There is also some evidence to suggest that phonologically disordered children may have poorer speech motor skills than their normally developing peers (e. g. Henry 1990, Edwards 1992, Waters 1992, Towne 1994). There are numerous techniques that can be used to measure a number of different aspects of speech motor control. However, there are methodological difficulties in devising appropriate protocols for the collection and analysis of speed of speech production as used as an index of speech motor skill in young children. Some of the techniques that have had clinical application include measurement of rate in connected speech production and measurement of diadochokinetic (DDK) repetition rate. This investigation compared normally developing and phonologically disordered preschool children on various indirect measures of speech motor skills, in imitated and spontaneous connected speech and in DDK tasks. The investigation also focused on refining the techniques of data collection and analysis appropriate to young children. While the results vary with regard to the statistical significance of the differences between the two groups of children on articulation rates and DDK rates, analysis of the error patterns in single word, spontaneous connected speech, imitated connected speech and DDK productions identified a sub group of phonologically disordered children who may present with an underlying speech motor deficit as the basis of their phonological disability. The results of the investigation are considered in terms of their implication for the speech motor skills of the two groups of children, techniques for measuring various aspects of speech motor skill and the clinical identification of phonologically disordered children who have an underlying speech motor deficit.

The fuel systems fitted to the current generation of civil transport aircraft are rather complicated, due to the presence of multiple tanks, pumps, valves and complex pipeline systems. During fuel transfer between the tanks, when controlling the aircraft centre of gravity or engine feed and refuel operations, a number of pumps and valves are involved resulting in complex pressure and flow interactions. In order to minimise the pressure surges during sudden system changes and flow overshoot during fuel transfer and refuelling, different motor drive system control strategies have been investigated. It is proposed that the current control method of electrically driven centrifugal-type pumps could be replaced by improved open and closed loop strategies where the flow overshoot can be minimised and pressure surges reduced. Steady-state and dynamic models of an AC induction motor drive and typical aircraft fuel system pipework components have been developed. The validation of these models has been performed using experimental data obtained from a fuel test rig constructed at the University of Bath using water as the working fluid. The simulation results have been shown to agree well with those from experimentation. In addition, the induction motor has been modelled based on its physical properties using the Finite Element Method software MEGA. The investigated fuel system has been described in linear terms and its behaviour has been identified. It is shown that the system dynamic behaviour can be controlled/improved using well established closed loop proportional-integral control. An open loop technique of simultaneous pump and valve control has been proposed and validated using experimental results, resulting in a reduction of both the transient pressure surges and flow overshoot during sudden valve closures, showing significant performance improvements. Improved closed loop control strategies for the pump drive system have also been developed in simulation. These are based on adaptive proportional-integral-derivative and fuzzy logic control strategies.

Wearable robotics for assistance and rehabilitation are not yet considered commercially mainstream products, and as a result have not yet seen advanced controls systems and interfaces. Consequently, the available technology is mostly adapted from systems used in parallel technologies, rather than custom applications intended for human use. This study concerns itself with the design and development of a custom control system for a 2-degree of freedom powered upper extremity orthosis capable of driving elbow flexion/extension 135º and humeral rotation 95º . The orthosis has been evaluated for use as both a long-term assistive technology device for persons with disabilities, and as a short-term rehabilitative tool for persons recovering injury. The target demographics for such a device vary in age, cognitive ability and physical function, thus requiring several input parameters requiring consideration. This study includes a full evaluation of the potential users of the device, as well as parameter considerations that are required during the design phase. The final control system is capable of driving each DOF independently or simultaneously, for a more realistic and natural coupled-motion, with proportional control by pulse-width modulation. The dual-axis joystick interface wirelessly transmits to the 1.21 pound control pack which houses a custom microcontroller-driven PCB and 1800 milliamp-hour lithium-ion rechargeable battery capable of delivering 4 hours of running time. Upon integration with the 2 DOF orthosis device, a user may complete full range of motion with up to 5 pounds in their hand in less than 7 seconds, providing full functionality to complete acts of daily living, thus improving quality of life.

We review some computational aspects of motor control. The problem of trajectory control is phrased in terms of an efficient representation of the operator connecting joint angles to joint torques. Efficient look-up table solutions of the inverse dynamics are related to some results on the decomposition of function of many variables. In a biological perspective, we emphasize the importance of the constraints coming from the properties of the biological hardware for determining the solution to the inverse dynamic problem.

Vision is important for the control of upper limb movements (Woodworth, 1899). How and when vision is used during a limb movement continues to be debated. In this thesis, I examined the use of visual feedback as a function of retinal limb position. Individuals made rapid upper limb aiming movements to a target location and vision was provided when the limb was at varying degrees of retinal eccentricity. The temporal characteristics, endpoint accuracy and precision, as well as the spatial variability of the limb trajectories were recorded and analyzed. No relationship was observed between retinal limb positions and the use of visual feedback during the movements. These results suggest that the use of vision during limb movements is not directly tied to the neuroanatomy of the eye and challenges continuous models of upper limb control.

Vision is important for the control of upper limb movements (Woodworth, 1899). How and when vision is used during a limb movement continues to be debated. In this thesis, I examined the use of visual feedback as a function of retinal limb position. Individuals made rapid upper limb aiming movements to a target location and vision was provided when the limb was at varying degrees of retinal eccentricity. The temporal characteristics, endpoint accuracy and precision, as well as the spatial variability of the limb trajectories were recorded and analyzed. No relationship was observed between retinal limb positions and the use of visual feedback during the movements. These results suggest that the use of vision during limb movements is not directly tied to the neuroanatomy of the eye and challenges continuous models of upper limb control.