Dissertations / Theses on the topic 'Army talks'

This thesis offers an alternative narrative why the Juba Peace Talks between the Government of Uganda and the rebel Lord’s Resistance Army (LRA) and its political wing, the Lord’s Resistance Movement (LRM), did not produce a Final Peace Agreement. Widely considered the most promising peace effort in the history of a violent conflict that began in 1986, talks were mediated by the Government of Southern Sudan from 2006 to 2008. During this time, the parties signed five separate agreements on a range of issues, yet in 2008 the LRA’s leader, Joseph Kony, failed to endorse them through a final signature. An aerial attack on the LRA by the Ugandan army spelled the end of the Juba Talks. It is commonly argued that as the first peace talks conducted with people wanted by the International Criminal Court, the Juba Talks collapsed because the arrest warrants made a negotiated agreement impossible. Another widely accepted reason is that the LRA/M were not committed to peace. This thesis, however, argues that how the LRA/M experienced the muddled and convoluted peace talks was the crucial factor because the dynamics of the process confirmed existing power dynamics. Internally, the LRA/M’s dynamics were profoundly influenced by their perception of being trapped in an established hostile system, causing a struggle to transform their own dynamics constructively. Offering an analytical chronology of the Juba Talks with an empirical emphasis on the perspective of the LRA/M and an analysis of LRA/M structures and behavioural patterns that emerged in the process, this thesis further outlines that judging success or failure of a peace process on whether agreements have been signed is misplaced. Despite not producing a final agreement, the Juba Talks contributed to peace and change in Uganda.

This thesis argues that the International Criminal Court (ICC) brings a new more deontological paradigm to international interventions, founded upon the universal application of legal principle, and displacing consequentialist notions of justice linked to human rights. Based upon the Court’s Statute and mode of operations, it is argued that this is associated with assumptions concerning the ICC’s primacy, military enforcement, and theory of change. The consequences of this development in volatile contexts are demonstrated. The case study, founded upon analysis from the war-affected community, examines the impact of the International Criminal Court in the Lord’s Resistance Army war, and reveals the relationship between criminal justice enforcement, and community priorities for peace and human rights. On the basis of evidence, and contrary to narratives repeated but unsubstantiated in the literature, it demonstrates that in this case these two imperatives were in opposition to one another. The Court’s pursuit of retributive legal principle was detrimental to the community’s interests in peace and human rights. The subsequent failure of the ICC’s review process to interrogate this important issue is also established. The research establishes that statutory and operational assumptions upon which Court interventions are based do not hold in volatile contexts. For the case study community and elsewhere, this has had adverse impacts, with significant implications for the ICC. The findings indicate that if these issues are not fundamentally addressed, principled international criminal justice enforcement in volatile environments will continue to have profoundly negative human rights consequences.

This thesis argues that the International Criminal Court (ICC) brings a new more deontological paradigm to international interventions, founded upon the universal application of legal principle, and displacing consequentialist notions of justice linked to human rights. Based upon the Court’s Statute and mode of operations, it is argued that this is associated with assumptions concerning the ICC’s primacy, military enforcement, and theory of change. The consequences of this development in volatile contexts are demonstrated. The case study, founded upon analysis from the war-affected community, examines the impact of the International Criminal Court in the Lord’s Resistance Army war, and reveals the relationship between criminal justice enforcement, and community priorities for peace and human rights. On the basis of evidence, and contrary to narratives repeated but unsubstantiated in the literature, it demonstrates that in this case these two imperatives were in opposition to one another. The Court’s pursuit of retributive legal principle was detrimental to the community’s interests in peace and human rights. The subsequent failure of the ICC’s review process to interrogate this important issue is also established. The research establishes that statutory and operational assumptions upon which Court interventions are based do not hold in volatile contexts. For the case study community and elsewhere, this has had adverse impacts, with significant implications for the ICC. The findings indicate that if these issues are not fundamentally addressed, principled international criminal justice enforcement in volatile environments will continue to have profoundly negative human rights consequences.

Thesis (M.S. in Information Operations)--Naval Postgraduate School, December 2008.<br>Thesis Advisor(s): Rothstein, Hy ; Jansen, Erik. "December 2008." Description based on title screen as viewed on February 5, 2009. Includes bibliographical references (p. 87-90). Also available in print.

The electromyographic (EMG) activity of trunk muscles has received little attention for tasks such as arm movements which do not explicitly direct trunk behavior. For such tasks, trunk muscles are considered to play a postural role, but only a limited set of conditions have been examined. The experiments presented in this study were designed to examine the relationship of the activity of trunk muscles at the initiation of rapid reaching movements to postural requirements as target direction and distance were broadly varied. Seated subjects performed bilateral arm reaching movements in the vertical plane to visible targets in many directions and at several distances. Surface EMG was recorded from trunk and shoulder muscles. Trunk position, trunk acceleration, and wrist acceleration were recorded. EMG patterns of the trunk muscles varied systematically with both target direction and target distance. The initial configuration of the body was varied to determine if the systematic variation with target direction was specific to one of three reference axes: the longitudinal axis of the forearm, the longitudinal axis of the trunk, or an absolute vertical axis that was external to the body. No one reference axis proved to have a stronger relationship with the activity of the trunk muscles across all target directions. Calculated muscle torques were compared to recorded activity of trunk muscles. Across all target directions and distances, trunk EMG at movement onset was not always qualitatively consistent with resisting either the calculated static torques of the final position or the segmental interactive effects of the arm on the trunk. By broadly varying target direction and distance, it was made clear that the activity of the trunk muscles was not universally related to any of the control requirements tested. Regardless, systematic changes in EMG patterns were observed for trunk muscles with target direction and target distance. The regularities of the data were more striking than the scatter. These regularities make the data a critical test for any proposed control schemes of trunk muscle activity.

The present thesis investigated the process of gross motor skill acquisition in the context of how movement pattern characteristics are organized and modified as a result of learning. The process of multisegmental skill acquisition was studied using performance production measures as a window to explore stable and transient changes in the movement control process. Bernstein's principles of exploitation of motion dependent torques, and the minimization of the degree of freedom were tested.<br>Twenty male subjects performed a total of 240 trials, as fast and accurate as possible, over 4 sessions for two tasks with different level of complexity (10 subjects per task). Kinematic and EMG data were acquired for a two segment model of the arm using a lightweight, adjustable manipulandum, and surface EMG for the PDL, PEC, TRI, and BIC muscles. Movement organization characteristics were revealed by wave pattern analyses performed with a phase breakdown technique, CCCF on torque partitioning components, and the CFs on the muscle activation waveforms.<br>The experimental protocol allowed enough practice for the tasks to be learned resulting in PT decrease. Consistent adaptations in movement organization were revealed by all three levels of analysis. Improvement in the performance production measures were in the same direction for both learning conditions, and the rate of change within experimental sessions decreased over time. The significant adaptations in movement organization occurred at the critical phases of each task (i.e. the power absorption--phase simple task, and the reversal phase--complex task). The EMG and joint torque components demonstrated significant amplitude and profile changes particularly during the critical phases. The subjects learned to perform the tasks faster by compensating the motion dependent torques components with the muscle torques in time and amplitude. This finding was consistent with the principle of exploitation of motion dependent torques which was enhanced with learning. Significant task specific motor program adaptations were revealed by the SVD analysis. The minimization of the degree of freedom hypothesis is consistent with the increasing covariation observed over learning between the antagonists PEC and BIC in the simple task, and for both the agonists PDL and TRI and antagonists PEC and BIC in the complex task. In addition the results justified the use of waveform analysis on performance production measures, as a method that can reveal important adaptation details in the underlying control mechanisms for gross motor skill.

Identification and development of physical characteristics that lead to efficient performance of military skills or tasks has been a consistently difficult task for the United States military for decades. The literature suggests certain physical characteristics may be more important, although this information is conflicting. Furthermore, the military physical fitness training program that is intended to prepare soldiers for combat is commonly evaluated with the Army Physical Fitness Test (PFT), a test that is more suited for evaluating health and wellness rather than task-specific fitness. All of this testing and training of soldiers focuses on the individual soldiers and their abilities although military skills or tasks are seldom if ever conducted independently. The first purpose of this dissertation was to identify relationships between the PFT, anthropometrics, measures of strength, and power. The second purpose was to identify the team characteristics that influence team performance during the Sandhurst Competition (a 2-day simulated military operation). Strong correlations were found between PFT events and weak correlations were found between PFT measures and evaluations of strength and power. The strong correlations between PFT events could indicate that only one event may be necessary to determine health and wellness. The weak correlations between events of the PFT and measures of strength and power suggest the PFT is not an assessment of strength and power based on the strength and power measures employed in the current study. The evaluation of team characteristics indicated that age (possibly experience) had the largest effect on Sandhurst Competition performance. Further analysis of each event supported the contention that age influences performance but also identified specific aerobic, anaerobic, and anthropometric variables that influenced performance on particular events. The data from this dissertation suggests that teams competing in the Sandhurst Competition should attempt to recruit team members with more experience, very high run scores, and high vertical jump heights.

L'approche classique pour le contrôle coopératif dans l’espace de travail d’un manipulateur bi-bras a été revisitée. En particulier, une nouvelle approche pour la formulation symétrique de la coordination bi-bras à l'aide du concept "virtual sticks" a été proposée à l'aide d'un torseur cinématique en utilisant des quaternions duaux. Le contrôle couplé dans l'espace de travail coopératif qui est proposé, i.e. le contrôle simultané, en position et en orientation, des points de consigne de l'espace de travail relatif et absolu a été comparé à un contrôleur proportionnel découplé traitant séparément les erreurs de position et d'orientation. Le contrôleur couplé a démontré un meilleur suivi de la pose et de l'orientation en termes de précision et de stabilité comparé au contrôleur découplé pour les tâches exigeant un fonctionnement plus rapide dans l'espace de travail relatif des manipulateurs bi-bras.L'approche de modélisation et de contrôle de l'espace de travail d’une tâche coopérative, en exploitant les torseurs cinématiques et des quaternions duaux, a été étendue à la modélisation de la coopération des doigts d'une main robotique anthropomorphique. De plus, le couplage des articulations des doigts sous-actionnés de la main robotique a été représenté à l’aide de la "jacobienne couplée" du doigt. La "jacobienne couplée" du doigt robotique a été utilisée pour le contrôle cinématique inverse, tout en lui permettant une intégration facile avec un bras robotique.L'idée d'un traitement couplé des variables en position et en orientation a été capitalisée en utilisant la conception d'une trajectoire de second ordre utilisant des quaternions duaux. Le contrôleur de trajectoire ainsi conçu est capable de suivre les points de consigne en pose en vitesse et en accélération, de l'effecteur en utilisant le modèle dynamique inverse du robot. Le contrôleur couplé en taux d’accélération résolue ("resolved rate acceleration") s'est avéré capable d'un contrôle de trajectoire plus précis, particulièrement en termes d'erreurs liées à l'orientation, que le contrôleur découplé classique qui traitait séparément les points de consigne en position et en orientation et ignorait l'effet de la rotation sur le mouvement de translation. De plus, cela a également permis de réduire les oscillations de la commande du couple des articulations lorsque le contrôleur a été implémenté pour le contrôle de l'un des bras du robot bi-bras Baxter.Enfin, un cadre complet pour la coordination des systèmes robotiques bi-bras a été proposé avec l'ajout d'un planificateur de tâches coopératives. La simplicité du torseur cinématique a également été exploitée pour la génération de trajectoires généralisées du second ordre pour des tâches nécessitant un mouvement simplifié, comme la translation, la rotation et la torsion autour d'un axe hélicoïdale arbitraire donné dans un repère connu. La méthode de génération de trajectoires a été étendue pour représenter les contraintes liées aux tâches impliquant un contact entre les objets en utilisant le concept de mécanisme virtuel<br>The classical approach for dual-arm cooperative task space control was revisited and the symmetric formulation of dual arm coordination using virtual sticks was implemented using screw-based kinematics with dual quaternion representation. The proposed coupled control of cooperative task space, i.e. simultaneous control of both position and orientation setpoints of relative and absolute task space was compared against the performance of a proportional decoupled controller treating position and orientation error separately. The coupled controller demonstrated better tracking of pose and orientation in terms of accuracy and stability compared to the decoupled controller for tasks requiring faster operation in the relative task space of dual-arm manipulators.The cooperative task space modelling and control approach using screw-based kinematics and dual quaternions were extended for the cooperation modelling of the fingers of an anthropomorphic robotic hand. Additionally, the coupling of joints in the underactuated fingers of the robotic hand was represented with a coupled finger Jacobian. The coupled Jacobian of the robotic finger was used for inverse kinematic control, while allowing easy integration with a robotic arm.The idea of coupled treatment of position and orientation variables was capitalized further with the design of a second-order trajectory tracker using dual quaternions. The trajectory controller hence designed was capable of tracking pose, velocity and acceleration setpoints for the end-effector using inverse dynamic model of the robot. The coupled resolved rate acceleration controller was found to be capable of tighter trajectory control, specially for error terms related to orientation, compared to the conventional decoupled controller that treated the position and orientation setpoints separately and ignored the inherent effect of rotation on translational motion. Additionally, it also led to lower oscillations in the joint torque command when implemented for the control of one of the arms of Baxter dual-arm robot.Finally, a complete framework for the coordination of bi-arm robotic systems was proposed with the addition of a cooperative task planner. The simplicity of screw theory was exploited additionally for parametrized generation of generalized second order trajectories for tasks requiring simplified motion, like translation, rotation and screw motion around an arbitrary 6D screw-axis given in a known reference frame. The trajectory generation method was extended to represent the constraints related to tasks involving contact between objects using the concept of virtual mechanism

The MURAB project, created and developed by the RAM group at the university of Twente in the Netherlands, aims to improve the precision and the efficiency of the breast biopsy procedure, reducing as much as possible the intervention time and the discomfort for the patient. In this framework, a lot of studies have been conducted concerning efficient control methods capable of positioning the end effector of a KUKA LBR Med robotic arm at the location where the hypothetical lesion has been individualized. Starting from these methods, the main goal of this Thesis is to develop a control algorithm able to modify the robot trajectory such that it can find and reach the lesion position, even in presence of variations of this target location; caused for example by small patient’s motions. After a brief literature review about the already existing robot assisted breast biopsy procedures, an elaborated study relative to the control strategies adopted in the collaboration and interaction between humans and robotic systems has been presented. This leads to the characterization and implementation of the final controller, defined as an impedance controller with energy and power limitations to increase the safety of the patient and energy tanks to ensure passivity. The validity and the stability of this control approach have been analyzed and tested firstly with a 20 sim simulation and then with experiments in laboratory. The results will show how it is possible to compensate rotational and / or translational errors in the robot end effector’s positioning through a suitable tuning of some parameters, while keeping a good image quality of the determined target and without loosing contact between the probe and the breast phantom / skin. Meanwhile, a proper characterization of the controller’s parameters allows to guarantee the compliance of the manipulator, fundamental to ensure the safety of the patient and of the interaction of the robot with the surrounding environment.

Surgeon expertise can be difficult to capture through direct robot programming. Deep imitation learning (DIL) is a popular method for teaching robots to autonomously execute tasks through learning from demonstrations. DIL approaches have been previously applied to surgical automation. However, previous approaches do not consider the full range of robot dexterous motion required in general surgical task, by leaving out tooltip rotation changes or modeling one robotic arm only. Hence, they are not directly applicable for tasks that require rotation and dual-arm collaboration such as debridement. We propose to address this limitation by formulating a DIL approach for the execution of dual-arm surgical tasks including changes in tooltip orientation, position and gripper actions.<br><br>In this thesis, a framework for multi-step surgical task automation is designed and implemented by leveraging deep imitation learning. The framework optimizes Recurrent Neural Networks (RNNs) for the execution of the whole surgical tasks while considering tooltip translations, rotations as well as gripper actions. The network architecture proposed implicitly optimizes for the interaction between two robotic arms as opposed to modeling each arm independently. The networks were trained directly from the human demonstrations and do not require to create task specific hand-crafted models or to manually segment the demonstrations.<br><br>The proposed framework was implemented and evaluated in simulation for two relevant surgical tasks, the peg transfer task and the surgical debridement. The tasks were tested under random initial conditions to challenge the robustness of the networks to generalize to variable settings. The performance of the framework was assessed using task and subtask success as well as a set of quantitative metrics. Experimental evaluation showed favorable results for automating surgical tasks under variable conditions for the surgical debridement, which obtained a task success rate comparable to the human task success. For the peg transfer task, the framework displayed moderate overall task success. Quantitative metrics indicate that the robot generated trajectories possess similar or better motion economy that the human demonstrations.

碩士<br>國立臺灣科技大學<br>材料科學與工程系<br>104<br>The universal popularity of industrial manipulators in a variety of industries can be observed due to the globalization of industries and the evolution of demographic structure. Following these dramatic changes, tasks, such as precision assembly, polishing and application of electronic components, which manipulators need to deal with in the future are becoming much more diversified. Position command-controlled robots in the past, therefore, can no longer achieve the tasks mentioned above. A position command-based impedance control with a assembly control strategy for implementation of screw assembly is exemplified in this thesis to present manipulators conducting precision assembly in position control mode. The use of impedance control equips the end effector on the manipulator with virtual mass, spring and damper, which enables the manipulator to complete the tasks in the unknown environments with the help of the end effector acquiring the measured data from the force/torque sensor and the contact force/torque of the exterior environments through the transformation of coordinate frame. The experiment in this thesis demonstrates how an end effector holds a screw component in the use of position command-based impedance control and screw assembly strategy to achieve the assembly task. Firstly, the necessity of the transformation of coordinate frame of the force/torque is to be proven and secondly the assembly task is to be implemented with position command-based impedance control and screw assembly strategy. Two types of screws (M8 and M6) are chosen in three separate tasks of different fastening torque to represent the abovementioned control method.

博士<br>國立臺灣科技大學<br>材料科學與工程系<br>104<br>As the technology of industrial robots develops and improves, industrial robots grow to execute the tasks interacted with environment, such as precise assembly tasks. Under the situation of positioning errors, using a pure position controller to control an industrial robot will cause the irregular impact, and destroy the industrial robot, tools, and the environment. The industrial robots are generally controlled under position mode. In order to utilize the industrial robots to execute the precise assembly task under position mode, this study proposes a strategy which named the variable impedance control with the desired force. This strategy is composed of the joint position based impedance control and the impedance parameters varying method. This strategy appropriately adjusts the position and orientation of the end-effector, and makes the robot compliant to the force/moment which acted by the environment. Using this strategy, the dynamics of the robot does not need to be derived. Therefore, this strategy can be utilized to control any position commanded robot. In addition, in order to correctly estimate external force/moment applied on the end-effector, this study derives the transformation of the force/moment on different coordinate frame. Using this transformation, the force/moment sensed from the coordinate frame of the sensor can be translated to the coordinate frame of the end-effector. The experiments of this study include the fundamental experiment of the transformation of force/moment and peg-in-hole assembly experiments. The fundamental experiment of transformation of force/moment verifies that the transformation of force/moment coordinate frame correctly transforms the reference coordinate frame of force/moment. Moreover, this study utilizes peg-in-hole assembly tasks to verify the impedance control with desired force and the impedance parameters varying method. The results show that the impedance controller with variable parameters is more suitable than one with fix parameters; compared with fix desired force, the variable desired force can decrease the process time of the assembly task. Finally, the proposed method is utilized to implement three peg-in-hole assembly tasks with the different sets of clearance.

碩士<br>高雄醫學大學<br>職業安全衛生研究所<br>98<br>Manual materials handling (MMH) is a component of many jobs and activities undertaken in life. It involves lifting, lowering, pushing, pulling and carrying objects by hand. It has been a popular theme in many research fields, especially manual lifting. The main popular reason is the devastating cost and human suffering caused by the severity of MMH related injuries. The arms lift tasks are necessary in our active daily living. One of the potential hazards in lifting is the related musculoskeletal injury of upper-extremity, especially the arms and shoulders. The objective of this study is to provide a way to avoid job&apos;&apos;s cumulative trauma disorders (CTD) through the work field and to provide the factory suggestion to improve the work environment. The Nordic Musculoskeletal Disorders Questionnaire (NMQ) and baseline risk identification ergonomic factors (BRIEF) were used to determine the risk factors of musculoskeletal disorders of the workspace. The portable electromyogram (EMG) was used to measure the activity of muscle groups from the certain factory during the" Material packing into a paper-bag “task. The questionnaire’s result shows that 88.9% of workers reported musculoskeletal discomfort after the job, and the highest prevalence rates were occurred in the segments of shoulder and wrist, followed by the elbow and knee. The result of BRIEF checklist reveals that neck and back were the high risk regions in all tasks. The result of portable EMG, no matter in males or females, the right brachioradialis appears the most evident muscle fatigue. Results showed that the repetitive arms lifts tasks have potential ergonomics hazards result in occurring musculoskeletal disorders. For decreasing and preventing the hazard risks, the recommendations are to enhance musculoskeletal disorders education training, to take more break times, to use the vacuum lifters make the handling of heavy containers easier and to adjust the height of worktable to personnel.

碩士<br>國立臺灣海洋大學<br>電機工程學系<br>105<br>The main purpose of this thesis is to design two bionic hexapod robots, one with an arm remotely controlled via Bluetooth by smart phone, and another with a camera having visual tracking capability. First, for the part of the bionic hexapod robot, find a more stable and efficient stepping way through gait analysis. Use the infrared remote control to send instructions to the Arduino Mega 2560 board which controls the servo motors to complete the basic movement of the robot gait. And then write a mobile phone app which uses Bluetooth module as a communication interface to control the hexapod robot to move forward, move backward, turn left, turn right and other actions, Further, combined with a robot arm, a bionic hexapod robot can be used to pick up objects. And then the combination of the image processing module with bionic hexapod robot equips the bionic robot with visual tracking capability.

碩士<br>南華大學<br>資訊管理學系<br>107<br>Military, as Constitution of Roc, should shoulder the duty of defending our homeland. In recent years, our military considers disaster relief as operation missions, which need to put equipment investment into. However, the new technology implementation project in Air Force has been influenced by organization culture, mission characteristic and member behavior. It is impacting on the effectiveness of new technology import projects and the implementation of disaster relief tasks. 　　This research probe into relationships between processes of blocked and bureaucratic organization's new technology implementation project, organization cultures, missions and members. Using case study method, this study sort out interview records and participant observations, and induct conclusions and perspectives, discovered culture of an organization and it influence to mission, member and the new technology implementation project, established relational integrated model between organization culture, mission, member and technology development project. Field of research including project management, interdisciplinary and multicultural, which provide other researchers a reference to advances project management processes of related organizations.