Search Results (1 - 25 of 103 Results)

Sort By  
Sort Dir
 
Results per page  

Smith, GregoryAn Analysis of Critically Enabling Technologies for Force and Power Limiting of Industrial Robotics
Master of Science (M.S.), University of Dayton, 2017, Electrical Engineering
The power and force limited (PFL) industrial robot market is one that is much underdeveloped and market demand is increasing every year. A literature review of critically enabling technologies for PFL robotics is conducted to evaluate the successes and limitations of developed and emerging PFL technologies. From this analysis a one link robot testbed is developed to test and gain a deep understanding of inherent torque sensing properties. Two custom sensing configurations and two custom torque plates are also designed to evaluate key torque sensing properties. Finally, an evaluation on these results lead to conclusions of the inherent effectiveness of the selected PFL enabling technologies.

Committee:

Raul Ordonez, Ph.D (Committee Chair); Russell Hardie, Ph.D (Committee Member); Chetan Kapoor, Ph.D (Committee Chair)

Subjects:

Robotics

Keywords:

Torque Sensor; Harmonic Drive; Capacitive Sensor; Optical Encoder; Power and Force Limited

Singh, AditiAUTOMATED DECLARATIVE GESTURE GENERATION FOR NON-EMOTIONAL HUMAN HUMANOID CONVERSATION
MS, Kent State University, 2017, College of Arts and Sciences / Department of Computer Science
Gesture is a visible but inaudible universal language for the communication of intent between two entities for communication. Gesture includes various timed postures and timed coordinated motions (voluntary and involuntary) of different body-parts such as head, neck, eyes, spine, hand, shoulders, torso and their combinations. It expresses both non-emotional moods, conversational modes and dynamic relationship between the speaker and the listener. It is one of three major components of human-human interaction along with facial expressions and speech (including silence). For humanoids to interact with humans in an acceptable meaningful manner, it has to understand conversation and respond in a naturalistic way mimicking human gestures. Human conversations exhibit many types of universally accepted gestures that involve coordinated combination of posture and movement of head, eye and jaw with different speed and frequency. This research describes various conversational gestures and develops a declarative modeling technique to express gestures as coordinated movement of organs that are translated to coordinated and synchronized movements of motors. These libraries of motor control instructions are interpreted using Python language that drives a combination of stepper and servo motors to emulate coordinated organ movements. The project has many advantages in providing company to elderly care and medical patient care, reading stories to children and providing personal secretary to people.

Committee:

Arvind K. Bansal (Advisor); Javed Khan (Committee Member); Cheng Chang Lu (Committee Member); Jong-Hoon Kim (Committee Member)

Subjects:

Robotics

Keywords:

conversation; data-driven; eye-movement; gesture; head-movement; humanoid; non-emotional; social robotics

Chen, ZhiangDeep-learning Approaches to Object Recognition from 3D Data
Master of Sciences, Case Western Reserve University, 2017, EMC - Mechanical Engineering
This thesis focuses on deep-learning approaches to recognition and pose estimation of graspable objects using depth information. Recognition and orientation detection from depth-only data is encoded by a carefully designed 2D descriptor from 3D point clouds. Deep-learning approaches are explored from two main directions: supervised learning and semi-supervised learning. The disadvantages of supervised learning approaches drive the exploration of unsupervised pretraining. By learning good representations embedded in early layers, subsequent layers can be trained faster and with better performance. An understanding of learning processes from a probabilistic perspective is concluded, and it paves the way for developing networks based on Bayesian models, including Variational Auto-Encoders. Exploitation of knowledge transfer--re-using parameters learned from alternative training data--is shown to be effective in the present application.

Committee:

Wyatt Newman, PhD (Advisor); M. Cenk Çavusoglu, PhD (Committee Member); Roger Quinn, PhD (Committee Member)

Subjects:

Computer Science; Medical Imaging; Nanoscience; Robotics

Keywords:

deep learning; 3D object recognition; semi-supervised learning; knowledge transfer

Bettaieb, Luc AlexandreA Deep Learning Approach To Coarse Robot Localization
Master of Sciences (Engineering), Case Western Reserve University, 2017, EECS - Electrical Engineering
This thesis explores the use of deep learning for robot localization with applications in re-localizing a mislocalized robot. Seed values for a localization algorithm are assigned based on the interpretation of images. A deep neural network was trained on images acquired in and associated with named regions. In application, the neural net was used to recognize a region based on camera input. By recognizing regions from the camera, the robot can be localized grossly, and subsequently refined with existing techniques. Explorations into different deep neural network topologies and solver types are discussed. A process for gathering training data, training the classifier, and deployment through a robot operating system (ROS) package is provided.

Committee:

Wyatt Newman (Advisor); Murat Cavusoglu (Committee Member); Gregory Lee (Committee Member)

Subjects:

Computer Science; Electrical Engineering; Robotics

Keywords:

robotics; localization; deep learning; neural networks; machine learning; state estimation; robots; robot; robot operating system; ROS; AMCL; monte carlo localization; particle filter; ConvNets; convolutional neural networks

Warner, Holly E.Optimal Design and Control of a Lower-Limb Prosthesis with Energy Regeneration
Master of Science in Mechanical Engineering, Cleveland State University, 2015, Washkewicz College of Engineering

The majority of amputations are of the lower limbs. This correlates to a particular need for lower-limb prostheses. Many common prosthesis designs are passive in nature, making them inefficient compared to the natural body. Recently as technology has progressed, interest in powered prostheses has expanded, seeking improved kinematics and kinetics for amputees. The current state of this art is described in this thesis, noting that most powered prosthesis designs do not consider integrating the knee and the ankle or energy exchange between these two joints. An energy regenerative, motorized prosthesis is proposed here to address this gap.

After preliminary data processing is discussed, three steps toward the realization of such a system are completed. First, the design, optimization, and evaluation of a knee joint actuator are presented. The final result is found to be consistently capable of energy regeneration across a single stride simulation. Secondly, because of the need for a prosthesis simulation structure mimicking the human system, a novel ground contact model in two dimensions is proposed. The contact model is validated against human reference data. Lastly, within simulation a control method combining two previously published prosthesis controllers is designed, optimized, and evaluated. Accurate tracking across all joints and ground reaction forces are generated, and the knee joint is shown to have human-like energy absorption characteristics. The successful completion of these three steps contributes toward the realization of an optimal combined knee-ankle prosthesis with energy regeneration.

Committee:

Daniel Simon, Ph.D. (Committee Chair); Hanz Richter, Ph.D. (Committee Member); Antonie van den Bogert, Ph.D. (Committee Member)

Subjects:

Biomedical Engineering; Engineering; Mechanical Engineering; Robotics

Desai, Viraj NAlgorithms for Needle-Tissue Interaction State Estimation in Robotic Surgical Suturing
Master of Sciences (Engineering), Case Western Reserve University, 2016, EMC - Mechanical Engineering
Autonomous execution of suturing in robotic surgery poses numerous challenges ranging from the difficulties associated with kinematics of the robotic arm to complexities in having a computational control to optimize the needle drive. In order to autonomously drive a suture needle through the tissue, a robot needs a lot of information including different types of forces acting on the suture needle, an estimate on the configuration of its needle grasp, parameters associated with the tissue and so on. Despite having precise motion control, and visual and force measurements, a robot needs information on the effects of its motion on the different needle-tissue interaction wrenches, and analyze this information in a control loop in order to follow a path as close to the ideal needle path plan as possible. Although deriving a robotic motion control algorithm is not within the scope of this research, estimating states and parameters involved in needle-tissue interaction would serve as foundation for development of a stable control, and is the essence of this research. These states include but are not limited to – tissue parameters of stiffness, friction and cutting; configuration of needle grasp; states relating to the position of the needle with respect to the tissue; and the compression wrench on the tissue due to needle insertion. This work is based on the model [1] which identifies the forces acting on the suture needle into three categories – friction force, cutting force and tissue compression force. Deriving these forces in a coordinate frame attached to the needle, helps make the equations more insightful, and using a sound mathematical approach they can be presented in a form that allows the robot to identify the different states and the change in magnitude of states with the controlled inputs, relating the state estimate to the measurement of total wrench at the sensor. Once the equations are derived, simulation and experimental validation is performed to access the practical feasibility of the concept. The state estimates obtained using the Unscented Kalman filter algorithm relate well to the actual values in simulation, and those obtained from parameter fitting to the experimental data.

Committee:

M. Cenk Cavusoglu (Advisor); Wyatt Newman (Committee Member); Roger Quinn (Committee Member)

Subjects:

Computer Science; Mechanical Engineering; Robotics

Keywords:

Robotics; Algorithms; Autonomous Suturing;

Bales, JohnDesign and Implementation of Two Undergraduate Laboratories Teaching PID Controller Design and Robotics Using Simulink and LEGO NXT
Master of Science (MS), Ohio University, 2014, Mechanical Engineering (Engineering and Technology)
This paper outlines the design and implementation of a novel system to design undergraduate laboratories with. An initial laboratory form was written for the Robotics and Control of Robotic Manipulators class utilizing a LEGO NXT system and the Simulink programming environment. This laboratory form is given to students and then revised after each student group based on answers to a questionnaire, direct observation, and direct questioning during the laboratory period. This process is repeated after each group. This subjective data was gathered and grouped into three groups for analysis. Grouping was used to showcase trends caused by changes made to the lab during the iteration process. The same process is utilized in the Linear Systems Control class. Recommendations are made for future laboratories that will feature the student designed controller from Linear Systems Control in the mandatory tasks assigned during the Robotics class. This allows the student to take previous work and build upon it, exposing them to the design process, and allows the University to receive a larger return on any investments it might make in LEGO NXT systems.

Committee:

Robert Williams, II, PhD (Advisor)

Subjects:

Education; Educational Theory; Engineering; Robotics

Keywords:

Robotics; LEGO NXT; Laboratory Design

Li, DingESA ExoMars Rover PanCam System Geometric Modeling and Evaluation
Doctor of Philosophy, The Ohio State University, 2015, Geodetic Science and Surveying
The ESA ExoMars rover, planned to be launched to the Martian surface in 2018, will carry a drill and a suite of instruments dedicated to exobiology and geochemistry research. To fulfill its scientific role, high-precision rover localization and topographic mapping will be important for traverse path planning, safe planetary surface operations and accurate embedding of scientific observations into a global spatial context. For such purposes, the ExoMars rover PanCam system will acquire an imagery network providing vision information for photogrammetric algorithms to localize the rover and generate 3-D mapping products. Since the design of the PanCam will influence the localization and mapping accuracy, quantitative error analysis of the PanCam design will improve scientists’ awareness of the achievable accuracy, and enable the PanCam design team to optimize the design for achieving higher localization and mapping accuracy. In addition, a prototype camera system that meets with the formalized PanCam specifications is also needed to demonstrate the attainable localization accuracy of the PanCam system over long-range traverses. Therefore, this research contains the following two goals. The first goal is to develop a rigorous mathematical model to estimate localization accuracy of this PanCam system based on photogrammetric principles and error propagation theory. The second goal is to assemble a PanCam prototype according to the system specifications and develop a complete vision-based rover localization method from camera calibration and image capture to obtain motion estimation and localization refinement. The vision-based rover localization method presented here is split into two stages: the visual odometry processing, which serves as the initial estimation of the rover’s movement, and the bundle adjustment technique, which further improves the localization through posterior refinement. A theoretical error analysis model for each of the localization stages has been established accordingly to simulate the rover localization error with respect to the traverse length. Additionally, a PanCam prototype was assembled with similar parameters as the latest technical specifications in order to systematically test and evaluate the ExoMars PanCam localization and mapping capabilities. The entire processing path from system assemblage, calibration, feature extraction and matching, as well as rover localization from field experiments has been performed in this research.

Committee:

Alper Yilmaz (Advisor); Alan Saalfeld (Committee Member); Ralph von Frese (Committee Member)

Subjects:

Geographic Information Science; Robotics

Keywords:

ExoMars; Localization; Error Propagation; Bundle Adjustment; Visual Odometry

Davis, Ronald JEVOLUTIONARY GROUND REACTION FORCE CONTROL OF A PROSTHETIC LEG TESTING ROBOT
Master of Science in Electrical Engineering, Cleveland State University, 2014, Washkewicz College of Engineering
Typical tests of prosthetic legs for transfemoral amputees prove to be cumbersome and tedious. These tests are burdened by acclimation time, lack of repeatability between subjects, and the use of complex gait analysis labs to collect data. To create a new method for prosthesis testing, we design and construct a robot that can simulate the motion of a human hip. We discuss the robot from concept to completion, including methods for modeling and control design. Two single-input-single-output (SISO) sliding mode controllers are developed using analytical and experimental methods. We use human gait data as reference inputs to the controller. When doing so we see the problems associated with the gait data that make it unfit for use as reference data. We apply a smoothing algorithm to correct the gait data. The robot is evaluated based on its ability to track the gait data. Despite proper tracking of the reference inputs, operating the robot with a passive prosthesis shows that the robot cannot adequately produce the ground reaction force (GRF) of an able bodied person. We devise a novel method to control GRF of the robot/prosthesis combination based on the way that human subjects walk with a prostheses. When walking with a prosthesis, users compensate for the deficiencies of the prosthesis by modifying their gait patterns. To simulate this we use an evolutionary algorithm called biogeography-based optimization (BBO). We use BBO to modify the reference inputs of the robot, minimizing the error between the able-bodied GRF data and that of the robot walking with the passive prosthesis. Experimental results show a 62% decrease in the GRF error, effectively showing the robot’s compensation for the prosthesis and improved control of GRF.

Committee:

Daniel Simon, Ph.D. (Committee Chair); Hanz Richter, Ph.D. (Committee Member); Antonie van den Bogert, Ph.D. (Committee Member); Eugenio Villaseca, Ph.D. (Committee Member)

Subjects:

Biomechanics; Biomedical Engineering; Electrical Engineering; Engineering; Robotics

Li, BorenHuman-like Robotic Handwriting and Drawing
Master of Science, The Ohio State University, 2012, Electrical and Computer Engineering
The method of human-like handwriting and drawing is addressed with a three-link arm. Three strategies of trajectory planning are considered: the basic stroke method, the Bezier Curve method and the non-gradient numerical optimization method. Scalar patterns are converted into vector form whose movement sequence and speed can be selected to imitate human handwriting and drawing. A nonlinear three-link three dimensional arm, similar to human arm, tracks the planned trajectories. The feasibility of these methods is demonstrated by simulation.

Committee:

Hooshang Hemami (Advisor); Yuan F. Zheng (Committee Member)

Subjects:

Robotics

Keywords:

Trajectory Planning; Three-Dimensional Three Link Humanoid Model; Handwriting; Drawing

MacRobbie, Danielle ElizabethAn Investigation of Technological Impressions in Steve Reich and Beryl Korot's Three Tales
Master of Music (MM), Bowling Green State University, 2013, Music History
The impact of technology upon the twentieth century and the influence it continues to exert upon the present human community is self-evident. The allure and power of technology are broadcast via the grandest media and performance entertainment, while on the opposite spectrum, technology is being continually refined to render its electro-mechanical or bio-technical feats for humans. It is this theme of the increasing growth and import of technology upon every facet of human life that serves as the subject of Three Tales, a twenty-first century documentary digital video opera by composer Steve Reich and video artist Beryl Korot. In this work, Reich and Korot confront society's negligence of particular directions that technological development and application have undergone in the past century, and advise against taking the same paths in the coming era. Even as modern technology is critiqued in Three Tales, the work itself bends to accept the reality of technology's significance upon modern thought and life. In keeping with Reich and Korot's categorization of the work as a "documentary digital video opera," Three Tales is a performance work heavily dependent upon technology for its generation, presentation, and discussion of the interchange between technology and humankind. This thesis will investigate how technology has shaped the course of an artwork whose purpose is to expose and debate the handling of technology in current society. Technology in Three Tales is examined from various perspectives. Chapter one presents the foundational role of technology as "tool," "subject," and "theme." Chapter two considers how visual and audio technologies are used in Three Tales to suggest the effects technology may have upon perceptions of human connectedness and isolation. Chapter three investigates the inherent paradox in Three Tales that occurs from using technological devices for the work's production while its theme critiques modern, technological advances. The chapter also considers the influence technology has upon the formation of Three Tales's generic identification.

Committee:

Eftychia Papanikolaou (Advisor); Alexa Woloshyn (Committee Member); Mary Natvig (Committee Member)

Subjects:

Biology; Ethics; History; Information Technology; Medical Ethics; Military History; Minority and Ethnic Groups; Music; Nanotechnology; Robotics; Robots; Spirituality; Technology; Theology

Keywords:

Steve Reich; Beryl Korot; Three Tales; Technology; Hindenburg zeppelin; Bikini Atoll; Cloning; electronic music; IRCAM; freeze frame sound; new music theater; Kismet; human connectedness; human isolation; technology and art; art and politics; paradox

Lonsberry, Alexander J.Fast Recognition and Pose Estimation for the Purpose of Bin-Picking Robotics
Master of Engineering, Case Western Reserve University, 2011, EMC - Mechanical Engineering
This thesis presents a novel object recognition engine for the application of bin-picking. The algorithm is capable of quickly recognizing and estimating the pose of objects in a given unorganized scene. Based on the oriented point-pair feature vector, the algorithm matches points in the scene to points on the surface of an original model via an efficient voting process. Groups of features defining a point in the scene are used to find probable matching model points in a precompiled database. Sets of candidate model and scene point-pair matches are created and then filtered based on a geometric consistency constraint. Results show that the algorithm can produce centroid error values of less than ≈.55mm and angular error values of less than ≈4° without a secondary iterative closest point algorithm. Run-times are in the range of .1 to .5 secs to locate a single object.

Committee:

Roger Quinn, PhD (Advisor); Frank Merat, PhD (Committee Member); Jaikrishnan Kadambi, PhD (Committee Member)

Subjects:

Robotics

Keywords:

Object Recognition; 3D; 3-D; pose estimation; automatic correlation; bin-picking; robotics; bin picking; vision system

JIANG, JINWEICollaborative Tracking of Image Features Based on Projective Invariance
Doctor of Philosophy, The Ohio State University, 2012, Geodetic Science and Surveying

In past manned lunar landing missions, such as Apollo 14, spatial disorientation of astronauts substantially compromised the productivities of astronauts, and caused safety and mission success problems. The non-GPS lunar environment has micro-gravity field, and lacks both spatial recognition cues and reference objects which are familiar to the human biological sensors related to spatial recognition (e.g. eyes). Such an environment causes misperceptions of the locations of astronauts and targets and their spatial relations, as well as misperceptions of the heading direction and travel distances of astronauts. These spatial disorientation effects can reduce productivity and cause life risks in lunar manned missions. A navigation system, which is capable of locating astronauts and tracking the movements of them on the lunar surface, is critical for future lunar manned missions where multiple astronauts will traverse more than 100km from the lander or the base station with the assistance from roving vehicle, and need real-time navigation support for effective collaborations among them.

Our earlier research to solve these problems dealt with developing techniques to enable a precise, flexible and reliable Lunar Astronaut Spatial Orientation and Information System (LASOIS) capable of delivering real-time navigation information to astronauts on the lunar surface. The LASOIS hardware was a sensor network composed of orbital, ground and on-suit sensors: the Lunar Reconnaissance Orbiter Camera (LROC), radio beacons, the on-suit cameras, and shoe-mounted Inertial Measurement Unit (IMU). The LASOIS software included efficient and robust algorithms for estimating trajectory from IMU signals, generating heading information from imagery acquired from on-suit cameras, and an Extended Kalman Filter (EKF) based approach for integrating these spatial information components to generate the trajectory of an astronaut with meter-level accuracy. Moreover, LASOIS emphasized multi-mode sensors for improving the flexibility and robustness of the system.

From the experimental results during three field tests for the LASOIS system, we observed that most of the errors in the image processing algorithm are caused by the incorrect feature tracking. This dissertation addresses the feature tracking problem in image sequences acquired from cameras. Despite many alternatives to feature tracking problem, iterative least squares solution solving the optical flow equation has been the most popular approach used by many in the field. This dissertation attempts to leverage the former efforts to enhance feature tracking methods by introducing a view geometric constraint to the tracking problem, which provides collaboration among features. In contrast to alternative geometry based methods, the proposed approach provides an online solution to optical flow estimation in a collaborative fashion by exploiting Horn and Schunck flow estimation regularized by view geometric constraints. Proposed collaborative tracker estimates the motion of a feature based on the geometry of the scene and how the other features are moving. Alternative to this approach, a new closed form solution to tracking that combines the image appearance with the view geometry is also introduced. We particularly use invariants in the projective coordinates and conjecture that the traditional appearance solution can be significantly improved using view geometry. The geometric constraint is introduced by defining a new optical flow equation which exploits the scene geometry from a set drawn from tracked features. At the end of each tracking loop the quality of the tracked features is judged using both appearance similarity and geometric consistency. Our experiments demonstrate robust tracking performance even when the features are occluded or they undergo appearance changes due to projective deformation of the template. The proposed collaborative tracking method is also tested in the visual navigation algorithm of the LASOIS system instead of original KLT tracking method for the experiment data from Moses Lake. The experimental analysis shows that the proposed collaborative tracking approach significantly improved the accuracy of the navigation solution.

Committee:

Alper Yilmaz, PhD (Advisor); Rongxing Li, PhD (Committee Member); Ralph Von Frese, PhD (Committee Member)

Subjects:

Computer Engineering; Computer Science; Geographic Information Science; Robotics

Keywords:

features; tracking; appearance similarity; projective geometry; projective invariants

Yetkin, HarunStabilzing Control of an Autonomous Bicycle
Master of Science, The Ohio State University, 2013, Electrical and Computer Engineering
In this thesis, we utilize the precession e fect of the gyroscope to stabilize the bicycle both at zero forward velocity and varying velocities. Equations of motion of a bicycle with a wheel mounted on its bottom are derived and a first order sliding mode controller is designed to achieve the goal of stabilization. In order to verify the designed feedback controller, two experimental setups are built; an inverted pendulum setup and a bicycle setup. SMC design for the static bicycle model is tested on both the inverted pendulum and the bicycle setups. In order to judge the performance of the controller, a well-tuned PID controller is also tested on these setups. Then, in the light of the experimental results obtained on the inverted pendulum setup, a controller scheme for the stabilizing control of an autonomous bicycle is designed and tested on various road structures through simulation environment.

Committee:

Umit Ozguner, Professor (Advisor); Keith Redmill, Professor (Committee Member)

Subjects:

Electrical Engineering; Engineering; Mechanical Engineering; Robotics; Robots

Keywords:

Gyroscopic stabilization; Unmanned bicycle stabilization; Autonomous bicycle

Singh, DaljeetPath Planning and Evolutionary Optimization of Wheeled Robots
Master of Science in Electrical Engineering, Cleveland State University, 2013, Fenn College of Engineering
Probabilistic roadmap methods (PRM) have been a well-known solution for solving motion planning problems where we have a fixed set of start and goal configurations in a workspace. We define a configuration space with static obstacles. We implement PRM to find a feasible path between start and goal for car-like robots. We further extend the concept of path planning by incorporating evolutionary optimization algorithms to tune the PRM parameters. The theory is demonstrated with simulations and experiments. Our results show that there is a significant improvement in the performance metrics of PRM after optimizing the PRM parameters using biogeography-based optimization, which is an evolutionary optimization algorithm. The performance metrics (namely path length, number of hops, number of loops and fail-rate) show 34.91%, 23.18%, 52.21% and 21.21% improvement after using optimized PRM parameters. We also experimentally demonstrate the application of path planning using PRM to mobile car-like robots.

Committee:

Dan Simon, PhD (Committee Chair); Nigamanth Sridhar, PhD (Committee Member); Chansu Yu, PhD (Committee Member)

Subjects:

Computer Engineering; Electrical Engineering; Engineering; Robotics

Keywords:

Proababilistic roadmap methods; Biogeography-based optimization; PRM; BBO; Path planning; Mobile robots

Colbrunn, Robb WilliamA Robotic Neuro-Musculoskeletal Simulator for Spine Research
Doctor of Engineering, Cleveland State University, 2013, Fenn College of Engineering
An influential conceptual framework advanced by Panjabi represents the living spine as a complex neuromusculoskeletal system whose biomechanical functioning is rather finely dependent upon the interactions among and between three principal subsystems: the passive musculoskeletal subsystem (osteoligamentous spine plus passive mechanical contributions of the muscles), the active musculoskeletal subsystem (muscles and tendons), and the neural and feedback subsystem (neural control centers and feedback elements such as mechanoreceptors located in the soft tissues) [1]. The interplay between subsystems readily encourages “thought experiments” of how pathologic changes in one subsystem might influence another—for example, prompting one to speculate how painful arthritic changes in the facet joints might affect the neuromuscular control of spinal movement. To answer clinical questions regarding the interplay between these subsystems the proper experimental tools and techniques are required. Traditional spine biomechanical experiments are able to provide comprehensive characterization of the structural properties of the osteoligamentous spine. However, these technologies do not incorporate a simulated neural feedback from neural elements, such as mechanoreceptors and nociceptors, into the control loop. Doing so enables the study of how this feedback—including pain-related— alters spinal loading and motion patterns. The first such development of this technology was successfully completed in this study and constitutes a Neuro-Musculoskeletal Simulator. A Neuro-Musculoskeletal Simulator has the potential to reduce the gap between bench and bedside by creating a new paradigm in estimating the outcome of spine pathologies or surgeries. The traditional paradigm is unable to estimate pain and is also unable to determine how the treatment, combined with the natural pain avoidance of the patient, would transfer the load to other structures and potentially increase the risk for other problems. The novel Neuro-Musculoskeletal Simulator described in this work has demonstrated, through simulation and cadaveric experimentation, that it is able to incorporate data from external sensors (e.g. force, motion tracking) to modulate spine biomechanical responses. In addition, the Neuro-Musculoskeletal Simulator exhibited the ability to use an estimated nociceptive response in unilateral facet arthritis to elucidate statistically significant compensatory kinetic and kinematic changes. These changes included a 37% increase in spine shear force, and an 18% increase in applied spine torque.

Committee:

Robert McLain, M.D. (Committee Chair); Antonie van den Bogert, Ph.D. (Committee Member); Lars Gilbertson, Ph.D. (Committee Member); Daniel Simon, Ph.D. (Committee Member); Michael Hammonds, Ph.D. (Committee Member); George Chatzimavroudis, Ph.D. (Committee Member)

Subjects:

Biomechanics; Biomedical Engineering; Biomedical Research; Engineering; Health Care; Kinesiology; Mechanical Engineering; Neurosciences; Robotics; Robots

Keywords:

Robot; Robotics; Musculoskeletal; Pain; Control; Spine; Cervical; Engineering; Neural; Cadaveric, Testing

Liu, TaomingA MAGNETICALLY-ACTUATED ROBOTIC CATHETER FOR ATRIAL FIBRILLATION ABLATION UNDER REAL-TIME MAGNETIC RESONANCE IMAGING GUIDANCE
Doctor of Philosophy, Case Western Reserve University, 2017, EECS - Electrical Engineering
This thesis focuses on design, modeling, and analysis of a magnetically actuated robotic intravascular catheter for performing atrial fibrillation ablation under magnetic resonance imaging guidance. Specifically: A three dimensional deflection model of a steerable catheter in free space is proposed and experimentally validated using a hardware prototype. In the proposed method, the catheter is modeled as a series of finite segments. For each finite segment, a quasi-static torque-deflection equilibrium equation is calculated using the beam theory. By using the deflection displacements and torsion angles, the kinematic model of the catheter is derived. A Jacobian-based iterative inverse kinematics method for controlling the steerable catheter is presented. The repeatability and accuracy of the open-loop control of the catheter system performing complex geometric trajectories using this inverse kinematics method is experimentally evaluated. The proposed three dimensional kinematic model is extended to incorporate the catheter-surface contact by taking contact forces and torques into account. A systematic approach to the design optimization of a magnetically-actuated steerable catheter for atrial fibrillation ablation in the left atrium, is proposed. The study investigates the relationship between the catheter material and the catheter's steering performance and evaluates the design optimization of the electromagnetic coils, such as the optimal winding turns for the coils, the optimal size for the side coils and the optimal locations of the coil sets on the catheter. The selected design is validated on a simulated atrial fibrillation ablation in a realistic left atrium model. The simulation verifies that the catheter is successfully able to reach every target on the circumferential lesions.

Committee:

Murat Cavusoglu, Dr. (Committee Chair); Wyatt Newman, Dr. (Committee Member); Mark Griswold, Dr. (Committee Member); Vira Chankong, Dr. (Committee Member); Francis Merat, Dr. (Committee Member)

Subjects:

Robotics; Robots

Keywords:

Robotic Catheter; Magnetically-Actuated Catheter; Atrial Fibrillation Ablation; Real-time MRI Guidance; Robotic Intravascular Catheter; Catheter Deflection Model; Iterative Inverse Kinematics; Catheter-Surface Contact Model; Catheter Design Optimization

Pabbu, Akhil SaiIncorporating Passive Compliance for Reduced Motor Loading During Legged Walking
Master of Science in Electrical Engineering (MSEE), Wright State University, 2017, Electrical Engineering
For purposes of travelling on all-terrains surfaces that are both uneven and discontinuous, legged robots have upper-hand over wheeled and tracked vehicles. The robot used in this thesis is a simulated hexapod with 3 degrees of freedom per leg. The main aim is to reduce the energy consumption of the system during walking by attaching a passive linear spring to each leg which will aid the motors and reduce the torque required while walking. Firstly, the ideal stiffness and location or the coordinates for mounting the spring is found out using gradient based algorithm called `Simultaneous Perturbation and Stochastic Approximation Algorithm’ (SPSA) on a flat terrain using data from a single walking step. Motor load is approximated by computing the torque impulse, which is the summation of the absolute value of the torque output for each joint during walking. Once the ideal spring and mount is found, the motor loading of the robot with the spring attached is observed and compared on three different terrains with the original loading without the spring. The analysis is made on a single middle leg of the robot, which is known to support the highest load when the alternating tripod gait is used. The obtained spring and mounting locations are applied to other legs to compute the overall energy savings of the system. Through this work, the torque impulse was decreased by 14 % on uneven terrain.

Committee:

Luther Palmer, III., Ph.D. (Advisor); Zach Fuchs, Ph.D. (Committee Member); Xiaodong Zhang, Ph.D. (Committee Member)

Subjects:

Electrical Engineering; Robotics

Keywords:

Legged robots; Energy optimization in legged robots; Optimization using SPSA; Gradient based optimization; Spring placement on a hexapod; Energy cost; Torque distribution

Hunt, Alexander JacobNeurologically Based Control for Quadruped Walking
Doctor of Philosophy, Case Western Reserve University, 2016, EMC - Mechanical Engineering
Current robotic control methods take advantage of high computing power to compute trajectories and perform optimal movements for a given task, yet these robots still fall far short of their animal counterparts when interacting with the environment. Animals dynamically adapt to varying terrain and small perturbations almost effortlessly. In order to improve our robotic systems and build better control methods, it makes sense to look more closely at how animals solve this interaction. In this work, I developed a control model of mammalian walking with models grounded in neuroscience and computational neuroscience. First, I developed a neuromechanical model of a rat with 14 degrees of freedom and 28 muscles, and I explored how hypothesized neural architectures can be used to coordinate four limbs in a walking gait for a rat. Additionally, through simulated ablation experiments, I developed hypotheses on how inter-leg pathways work together to maintain limb timing. After this, I developed a procedure to train the neural system to produce dynamic walking in both a rat simulation and a robot named Puppy. This method works by first using a model of the system (rat or robot) to determine required motor neuron activations to produce stable walking. For the robot, this required building a force-length-pressure model of the McKibben actuators to enable accurate force control. Parameters in the neural system are then set such that it produces similar activations to the desired pattern. I applied the same training procedure to both the simulated rat and the robot and show that it is capable of producing continuous, self-supported stepping in both systems.

Committee:

Roger Quinn, PhD (Advisor); Joseph Mansour, PhD (Committee Member); Kiju Lee, PhD (Committee Member); Hillel Chiel, PhD (Committee Member)

Subjects:

Mechanical Engineering; Neurosciences; Robotics; Robots

Keywords:

Robotics; Computational Neuroscience; McKibben; Rat; Dog; Control; Festo; CPG; Afferent Feedback

Renfrew, Mark E.Algorithms for Active localization and Tracking in Image-Guided Robotic Surgery Systems
Doctor of Philosophy, Case Western Reserve University, 2016, EECS - Computer Engineering
Robotic surgery is an emerging technology that promises great improvements in surgical outcomes and reductions in post-surgical recovery time for patients. Although several commercial and research robotic surgical systems have been demonstrated, many challenges remain to be overcome before surgical robotics fully realize their lifesaving potential. This dissertation details several approaches to improve the state of the art in the field. The first contribution of this work is to describe a framework of algorithms for the active localization and tracking of flexible needles and targets during image-guided percutaneous interventions. The needle and target configurations are tracked by Bayesian filters employing models of the needle and target motions and measurements of the current system state obtained from an intra-operative imaging system which is controlled by an entropy-minimizing active localization algorithm. Versions of the system were built using particle and unscented Kalman filters and their performance was measured using both simulations and hardware experiments with real magnetic resonance imaging data of needle insertions into gel phantoms. Performance of the localization algorithms is given in terms of accuracy of the predictions and computational efficiency is discussed. The second contribution of this work is a level set-based method for the segmentation of biological images. Regions of interest are grown and find the natural boundaries in the volume. Seeding is done either manually or by an automatic recursive variance-based process. The procedure is run on an MR image volume of a cancer patient and demonstrated to successfully segment the data.

Committee:

Cenk Cavusoglu, PhD (Committee Chair); Frank Merat, PhD (Committee Member); Wyatt Newman, PhD (Committee Member); Greg Lee , PhD (Committee Member); Kiju Lee, PhD (Committee Member)

Subjects:

Computer Science; Robotics

Sridhar, Dheerendra M.Mathematical Modeling of Cable Sag, Kinematics, Statics, and Optimization of a Cable Robot
Master of Science (MS), Ohio University, 2015, Mechanical Engineering (Engineering and Technology)
Cable sag can have significant effects on the cable length computation in a cable robot and this is more pronounced in large scale cable robots, such as the Algae Harvesting Cable Robot System. This requires modeling the cable as a catenary instead of an approximated straight line model. Furthermore, when there is actuation redundancy involved, the modeling and simulation of the system becomes much more complex, requiring optimizing routines to solve the problem. The cable sag compensated or the catenary model was used for the Algae Harvesting Cable Robot System and simulated to solve the Kinematics and Statics problems. This involved optimization of cable tensions and finding the errors involved in the cable length. A relative comparative analysis between the straight line and cable sag model is presented. Finally based on the qualitative and quantitative results obtained, recommendations were made on the choice of model and solution methodologies.

Committee:

Robert Williams, II (Advisor); Hajrudin Pasic (Committee Member); Greg Kremer (Committee Member); Vardges Melkonian (Committee Member)

Subjects:

Applied Mathematics; Engineering; Mechanical Engineering; Robotics

Keywords:

Mechanical Engineering; Robotics; Cable Robot; Cable Sag; Tension; Mathematical Modeling; Optimization; Kinematics; Statics

Bin Hammam, Ghassan MohammedWhole-Body Motion Retargeting for Humanoids
Doctor of Philosophy, The Ohio State University, 2014, Electrical and Computer Engineering
Humanoid motion generation using human-to-humanoid motion transfer (motion retargeting) has become an essential and useful method to produce human-like motions. However, this method has a number of challenges that limit its usefulness. The objective of this dissertation is to develop efficient methods to retarget whole-body, task-space motion from a human to a humanoid while managing online (real-time) dynamic (balance) and kinematic (joint-limit, self-collision, and foot) constraints, and while ensuring motion tracking characterized by dynamic consistency with natural human motion. This dissertation describes a very efficient, modified resolved acceleration control (MRAC) algorithm for dynamic filtering and control of whole-body humanoid motion in response to upper-body task specifications, or a commanded joint-space motion reference in general. The dynamic filter is applicable for general motions when standing in place. It is characterized by modification of the commanded torso acceleration based on a geometric solution to produce a Zero Moment Point (ZMP) which is inside the foot support. The resulting feasible, modified motion is synchronized to the reference motion when the computed ZMP for the reference motion again falls within the support. Contact forces at each foot are controlled through a dedicated force distribution module which optimizes the ankle roll and pitch torques. MRAC uses time-local information and is therefore targeted for online control. The effectiveness of the algorithm is demonstrated by means of simulated experiments. This dissertation presents a Cartesian-space constrained resolved acceleration control (CRAC) framework to manage execution of operational motion-tracking tasks, and handle constraints for redundant and non-redundant task specifications. The approach is particularly well suited for online control of humanoid robots using captured human motion data expressed by Cartesian variables. The current formulation is dynamically consistent, and enforces kinematic constraints such as joint-limit, self-collision, and foot constraints. Based on the applied reference motion and constraints setup, CRAC enables motion control at rates of 3k Hz in the fastest case and more than 400 Hz in the slowest case. A method, called the Unified CRAC (UCRAC), that handles dynamic balance and kinematic constraints together for whole-body task-space motion is also developed. The approach is non-iterative and as a result suitable for real-time applications. It utilizes an efficient centroidal dynamics formulation to relate the net force, applied to the humanoid by the environment, to the joint accelerations that realize the motion. UCRAC uses heuristically-designed rules to force the computed ZMP and centroidal projection to remain inside the foot support area, and to compute and command the net force on the system. Enforcing foot constraints are managed through the application of the constrained centroidal dynamic equations. UCRAC is capable to handle quasi-static tic balance in addition to dynamic balance, which is shown to provide significant flexibility for different humanoid foot dimensions. This method performs with rates providing faster than real-time performance with a minimum speed of 250 Hz which includes all constraint computations as well. The efficacy of the proposed methods is demonstrated in all cases by simulated and real-time experiments of task-level human motion replication on a Honda-like humanoid robot model.

Committee:

David Orin, Prof. (Advisor); Yuan Zheng, Prof. (Committee Member); Kevin Passino, Prof. (Committee Member)

Subjects:

Computer Engineering; Electrical Engineering; Robotics

Keywords:

Humanoid; Motion retargeting; centroidal dynamics; human-like motions; dynamic constraints; balance; kinematic constraints; joint-limit; self-collision; Cartesian-space controller; resolved acceleration control; dynamically consistent

Smith, Lauren MelissaThe Tri-Wheel: A Novel Robot Locomotion Concept Meeting the Need for Increased Speed and Climbing Capability
Master of Sciences, Case Western Reserve University, 2015, EMC - Mechanical Engineering
A need has been expressed for a robot locomotion concept that incorporates both efficient, rapid motion on smooth surfaces as well as the capacity to traverse a variety of challenging terrain obstacles, including but not limited to: stairs, rubble, and other environmental impediments. Currently, this dual capability has not been optimized successfully for existing locomotion concepts. This research seeks to meet this need with a novel mobility concept called the Tri-Wheel and chronicles its theoretical conception, design, and preliminary testing. An in-depth discussion of the design process and determination of derived requirements is first presented to substantiate the final configuration. The Tri-Wheel is then manufactured and installed on an existing robot chassis for testing, ultimately proving the concept successful by meeting the stated research objectives.

Committee:

Roger Quinn (Committee Chair); Paul Barnhart (Committee Member); Joseph Mansour (Committee Member)

Subjects:

Engineering; Mechanical Engineering; Robotics; Robots

Keywords:

robot locomotion; wheels; gearing; first responder; novel locomotion platform; wheel-leg hybrid; Tri-Wheel

Hughes, Bradley EvanA Navigation Subsystem for an Autonomous Robot Lawn Mower
Master of Sciences (Engineering), Case Western Reserve University, 2011, EECS - Electrical Engineering
This thesis describes a cost effective, accurate, and precise electronic navigation system which is suitable for outdoor commercial mobile robots. The hardware design of the system incorporates commercial off the shelf Global Positioning System receiver modules and support electronics. The software design of the system makes use of an open source positioning library to enable Real Time Kinematic satellite positioning. The designed navigation system has been integrated with a preexisting mobile robot platform, an autonomous robot lawn mower, which includes a set of reference sensors to provide accurate robot pose information. The reference platform is used to quantitatively evaluate the performance of the new cost effective system. A degradation factor of 1.7 in terms of positional accuracy is traded off in favor of achieving a cost savings factor of about thirty.

Committee:

Roger Quinn, PhD (Advisor); Roger Quinn, PhD (Committee Chair); Marc Buchner, PhD (Committee Member); Francis Merat, PhD (Committee Member)

Subjects:

Computer Engineering; Electrical Engineering; Robotics; Robots

Keywords:

mobile robot; autonomous guided vehicle; global positioning system; navigation system; control system; robot lawn mower; consumer product automation; low cost navigation

Karadogan, ErnurA Cable-Actuated Robotic Lumbar Spine as the Haptic Interface for Palpatory Training of Medical Students
Doctor of Philosophy (PhD), Ohio University, 2011, Mechanical Engineering (Engineering and Technology)

The role of simulation in medical education is rapidly increasing. Simulations to train nurses, veterinarians and doctors (osteopathic and allopathic) are utilized due to their effectiveness and cost-reducing advantages. These simulations can be computer-based or in the form of mannequins that can simulate some functions of the real human body such as breathing, blood pressure, pulse and temperature, among others. Computer-based haptic simulations require the usage of a haptic interface to interact with virtual objects. That is clearly not the case when humans interact with real objects. Therefore, a system which allows the user to interact with a real object could be a more realistic and effective approach.

This dissertation presents the theoretical framework (kinematics, pseudostatics, dynamics and control) of a novel 15 degree-of-freedom cable-actuated robotic lumbar spine (RLS) which can mimic in vivo human lumbar spine movements to provide better hands-on training for medical students. The design incorporates five active lumbar vertebrae and the sacrum, with dimensions of an average adult human spine. It is actuated by 20 cables connected to electric motors. Every vertebra is connected to the neighboring vertebrae by spherical joints. The RLS is designed to be controlled by a force-feedback joystick or an affordable haptic device. By moving the joystick, the angles of rotations are commanded to the RLS, therefore representing a normal lumbar spine movement. A static model of the human lumbar spine was also derived to obtain these normal movement patterns for different types of motion.

Medical schools can benefit from a system that will help instructors train students and assess their palpatory proficiency throughout their education. The RLS has the potential to support these needs in palpatory diagnosis. Medical students will be given the opportunity to examine their own patient that can be programmed with a variety of dysfunctions related to the lumbar spine before they start their professional lives as doctors. The robotic lumbar spine can be used to teach and test medical students to be able to recognize normal and abnormal movement patterns of the human lumbar spine under flexion, extension, lateral bending and axial torsion.

Committee:

Robert Williams, PhD (Advisor); John N. Howell, PhD (Committee Member); Hajrudin Pasic, PhD (Committee Member); Gursel Suer, PhD (Committee Member); Gary Chleboun, PhD (Committee Member)

Subjects:

Biomechanics; Mechanical Engineering; Robotics

Keywords:

Cable robot; human lumbar spine; Palpatory training; RLS; Robotic lumbar spine; three-dimensional static modeling

Next Page