Some relevant aspects of dynamics and control of heavy-duty hydraulic machines in a teleoperated mode were investigated. These machines, such as excavators and forest harvesters, are mostly used in primary industries. They have a manipulator-like structure with a nonlinear and coupled actuating system. The aim of the project is to investigate different approaches towards converting such machines, with minimum changes, into task-oriented human-supervisory control systems. This provides the opportunity to use both human supervision and robotic power in hazardous environments and for tasks for which human decision is necessary. A methodology was developed for fast and accurate simulations. Analytical, steady-state and numerical techniques were combined using Large-Scale Systems analysis. The inclusion of nonlinearities in the form of discontinuities (e.g., gear backlash and stick-slip friction) in the model was investigated. Numerical simplifications of the structural dynamics and alternative solutions for the hydraulic part were also studied. The model describing the performance of the machine has been written in ACSL (Advanced Continuous Simulation Language) on a VAX computer system. A modified version of the program is at present running close to real-time on a single processor in conjunction with high speed graphics in a manner similar to a flight simulator used for human interface studies and training. The model also evaluates the performance of the machine in a teleoperated mode and under different control strategies. As a result a velocity control algorithm has been developed which is applied in conjunction with the closed-loop components for teleoperation of heavy-duty hydraulic machines; it is basically a feedforward compensation which uses the measured hydraulic line pressures along with fluid-flow equations as criteria to control the joint velocities as well as to uncouple the interconnected actuating system. The control algorithm has been written in C language and is running on an IRONICS computer system, interfaced between the human operator and the machine. The simulation results are supported by the experimental evidence. The experiments were performed on a Caterpillar 215B excavator. Improved operator safety, extension of human capability, job quality and productivity increase are the advantages of a successful implementation of robotic technology to these industrial machines. / Applied Science, Faculty of / Mechanical Engineering, Department of / Graduate

Over the last several years a branch of Artificial Intelligence called Qualitative Reasoning has received much attention. A qualitative reasoner use qualitative values such as increasing, boiling and turbulent to analyze the behavior of physical systems. Existing qualitative frameworks have focused on physical systems whose qualitative values can be identified given the value of a single parameter. This precludes the application of qualitative models to physical systems whose properties require the values of several parameters. An example of such a system is the kinematics of a robotic manipulator. With this motivation, this thesis answers the following: What is a Qualitative model? Although current approaches appear diverse, they share a common mathematical foundation. This foundation is used to reformulate the qualitative model as a set of equivalence relations. The other question answered is: What extensions are needed to handle multivariate properties such as those encountered in the manipulator paradigm? The equivalence classes associated with qualitative models are geometrically shown to be connected hyperspaces. We show that existing frameworks are limited in the types of hyperspaces they can represent. The major ideas in this thesis are illustrated using manipulator kinematics. / Science, Faculty of / Computer Science, Department of / Graduate

No description available.

This thesis explores the possibility of using mechanical control in the design of a complex end effector. A design methodology is developed and demonstrated. The main goal of this methodology is to maximize reversible steps to direct the design. By attempting to obtain as much mechanical control as possible, several mechanisms are developed which could be used in applications where control of multiple operations by one motor is desired. Along with the demonstration of the design methodology with an end effector design, the application of this methodology to cigarette packaging machines is discussed. / Master of Science

In this thesis, forward and inverse kinematic equations are developed for a parallel, closed-loop manipulator known as the Variable Geometry Truss or VGT for short. Widely recognized as adaptive or collapsing structures for space and military applications, VGTs have not received due consideration as robotic manipulators. VGTs undoubtedly represent an important sector of future manipulator applications. VGTs are typically constructed using repeating identical cells or modules and they have exceptional stiffness to weight ratios. The data obtained from solving the forward kinematic equations is used for animation of the VGT. For animation, three dimensional graphics software, graPHIGS is used. Additionally, the kinematic analysis equations are used to map out workspace of the VGT. An experiment is also carried out to verify the computational results. / Master of Science

This work bridges the gap between theory and practice. The development of general inverse kinematic solution techniques is new, hence few detailed applications of these methods exist. Before methods such as these were available, most commercial manipulators were designed to be geometrically simple, yielding 4th or lower degree governing equations. With the further development and application of these techniques, industry will be capable of implementing more complex manipulators for highly specialized tasks. A general inverse kinematic analysis technique is applied to an industrial manipulator designed for the inspection of nuclear reactor vessels. The analysis is performed by solving the 16th degree univariate displacement polynomial of the general six-degree-of-freedom arm using an equivalent seven-degree-of-freedom closed-loop spatial chain. All possible combinations of joint angles for a given hand position and orientation are obtained. A region in which the manipulator has the maximum number of solutions is used as a numerical example. The inverse kinematic analysis was programmed in C, which is included in Appendix D. / Master of Science

The focus of this work is the real-time prediction and compensation of static deflections in robotic manipulator arms. A general manipulator deflection model is developed based on static beam theory and robot kinematics. An optimization technique is proposed to determine the stiffness of the manipulator components using end-effector deflection measurements. Strategies for incorporating this modeling approach into a manipulator controller are also presented along with the results of a successful application of this research. This work is an extension of previous manipulator deflection research. Multiple pairs of torsional stiffness elements and beam elements are used to model complex link and joint geometries whereas previous models only used a single beam per manipulator link. In addition, the modeling algorithms and stiffness characterization methods are general and may be applied directly to any serial manipulator. Also, the optimization techniques used to characterize a manipulator's stiffness provide a more accurate and flexible approach than previous analytical methods. The deflection model was successfully tested using a nuclear steam generator service manipulator. Since this manipulator is considerably more flexible than common industrial robots, it serves as a near worst-case test for deflection modeling. The end effector was found to deflect as much as 1.5 inches due to the weight of the links and joints. The deflection model was able to compensate for 94% of the end-effector deflection, allowing the manipulator to perform tasks requiring a positioning accuracy of 0.09 inches. The algorithms for flexible forward and inverse kinematics as well as trajectory generation were incorporated directly into the manipulator controller code. These modules were capable of running in real-time with little computational expense. / Master of Science

No description available.

The growing need in industrial applications of high-performance robots has led to designs of lightweight robot arms. However the light-weight robot arm introduces accuracy and vibration problems. The classical robot design and control method based on the rigid body assumption is no longer satisfactory for the light-weight manipulators. The effects of flexibility of light-weight manipulators have been an active research area in recent years. A new approach to correct the quasi-static position and orientation error of the end-effector of a manipulator with flexible links is studied in this project. In this approach, strain gages are used to monitor the elastic reactions of the flexible links due to the weight of the manipulator and the payload in real time, the errors are then compensated on-line by a control algorithm. Although this approach is designed to work for general loading conditions, only the bending deflection in a plane is investigated in detail. It is found that a minimum of two strain gages per link are needed to monitor the deflection of a robot arm subjected to bending. A mathematical model relating the deflections and strains is developed using Castigliano's theorem of least work. The parameters of the governing equations are obtained using the identification method. With the identification method, the geometric details of the robot arms and the carrying load need not be known. The deflections monitored by strain gages are fed back to the kinematic model of the manipulator to find the position and orientation of the end-effector of the manipulator. A control algorithm is developed to compensate the deflections. The inverse kinematics that includes deflections as variables is solved in closed form. If the deflections at target position are known, this inverse kinematics will generate the exact joint command for the flexible manipulator. However the deflections of the robot arms at the target position are unknown ahead of time, the current deflections at each sampling time are used to predict the deflections at target position and the joint command is modified until the required accuracy is obtained. An experiment is set up to verify the mathematical model relating the strains to the deflections. The results of the experiment show good agreement with the model. The compensation control algorithm is first simulated in a computer program. The simulation also shows good convergence. An experimental manipulator with two flexible links is built to prove this approach. The experimental results show that this compensation control improves the position accuracy of the flexible manipulator significantly. The following are the brief advantages of this approach: the deflections can be monitored without measuring the payload directly and without the detailed knowledge of link geometry~ the manipulator calibrates itself with minimum human intervention; the compensation control algorithm can be easily integrated with the existing uncompensated rigid-body algorithm~ it is inexpensive and practical for implementation to manipulators installed in workplaces.

Deploying robots in real-world environments involves immense engineering complexity, potentially surpassing the resources required for autonomous vehicles due to the increased dimensionality and task variety. To maximize the chances of successful real-world deployment, finding a simple solution that minimizes engineering complexity at every level, from hardware to algorithm to operations, is crucial. In this dissertation, we consider a vision-based manipulation system that can be deployed in-the-wild when trained to imitate sufficient quantity and diversity of human demonstration data on the desired task. At deployment time, the robot is driven by a single diffusion-based visuomotor policy, with raw RGB images as input and robot end-effector pose as output. Compared to existing policy representations, Diffusion Policy handles multimodal action distributions gracefully, being scalable to high-dimensional action spaces and exhibiting impressive training stability. These properties allow a single software system to be used for multiple tasks, with data collected by multiple demonstrators, deployed to multiple robot embodiments, and without significant hyper-parameter tuning. We developed a Universal Manipulation Interface (UMI), a portable, low-cost, and information-rich data collection system to enable direct manipulation skill learning from in-the-wild human demonstrations. UMI provides an intuitive interface for non-expert users by using hand-held grippers with mounted GoPro cameras. Compared to existing robotic data collection systems, UMI enables robotic data collection without needing a robot, drastically reducing the engineering and operational complexity. Trained with UMI data, the resulting diffusion policies can be deployed across multiple robot platforms in unseen environments for novel objects and to complete dynamic, bimanual, precise, and long-horizon tasks. The Diffusion Policy and UMI combination provides a simple full-stack solution to many manipulation problems. The turn-around time of building a single-task manipulation system (such as object tossing and cloth folding) can be reduced from a few months to a few days.