Browsing by Subject "Robotics"
Now showing 1 - 20 of 52
Results Per Page
Sort Options
Item A methodology to improve the cooperative performance of hedonistic multi-agents(Texas Tech University, 2006-12) Helm, Michael T.; Cooke, Daniel E.; Becker, Klaus G.; Pyeatt, Larry D.; Rushton, J. NelsonMulti-agent systems have several performance advantages over monolithic systems, among them are: improved speed performance, spatial distribution, reduction of single point failures, and both diversity and redundancy of capability. Such systems are often difficult to design and manage due to the complexity of coordination. The communications overhead cost to achieve coordination becomes intractable as the number of agents grows large. Yet, cooperative behavior has been shown to emerge from the interactions of simple multi-agents that use self-serving fixed-action patterns. Social insects are the inspiration for this work, which explores the value of constrained communication in simple hedonistic multi-agents for the purpose of improving their summed performance from the total system level view. The objective of this work is to develop a better understanding of the influence of communication in achieving improved cooperative system level performance among self-serving agents. This work includes development of an agent/world model and three experimental studies, the last one of which focuses on the cost/value of communications.Item A Scalable Framework for Parallelizing Sampling-Based Motion Planning Algorithms(2014-04-29) Jacobs, Samson AdeMotion planning is defined as the problem of finding a valid path taking a robot (or any movable object) from a given start configuration to a goal configuration in an environment. While motion planning has its roots in robotics, it now finds application in many other areas of scientific computing such as protein folding, drug design, virtual prototyping, computer-aided design (CAD), and computer animation. These new areas test the limits of the best sequential planners available, motivating the need for methods that can exploit parallel processing. This dissertation focuses on the design and implementation of a generic and scalable framework for parallelizing motion planning algorithms. In particular, we focus on sampling-based motion planning algorithms which are considered to be the state-of-the-art. Our work covers the two broad classes of sampling-based motion planning algorithms--the graph-based and the tree-based methods. Central to our approach is the subdivision of the planning space into regions. These regions represent sub- problems that can be processed in parallel. Solutions to the sub-problems are later combined to form a solution to the entire problem. By subdividing the planning space and restricting the locality of connection attempts to adjacent regions, we reduce the work and inter-processor communication associated with nearest neighbor calculation, a critical bottleneck for scalability in existing parallel motion planning methods. We also describe how load balancing strategies can be applied in complex environments. We present experimental results that scale to thousands of processors on different massively parallel machines for a range of motion planning problems.Item Application of the matrix approach to the kinematic modeling and analysis of spaial mechanisms(Texas Tech University, 1995-05) Suryanarayan, Krishna PrasadThe purpose of this thesis is to apply the homogeneous matrix transformation approach to the design of automotive alignment mechanisms and to briefly review its application to the kinematic analysis of a hyper-redundant manipulator. The homogeneous matrix transformation approach allows the determination of the position, velocity and acceleration of any point on the mechanism, defined in local coordinates, with respect to the world coordinate system, from a knowledge of the joint and link parameters. This homogeneous matrix method presents itself as a unique tool in the analysis of automotive alignment mechanisms as currently, these mechanisms are designed using geometric methods which involve tedious calculations. By the application of this matrix approach to their analysis, the design process is greatly simplified and it enables the design of these mechanisms to be easily implemented by breaking the design into smaller modules which are then solved individually to yield the final solution. Additionally, an investigation is performed to show the use of this matrix approach in the study of the kinematics of a hyper-redundant manipulator which is composed of wedge shaped discs.Item Automated conceptual design of manufacturing workcells in radioactive environments(2013-08) Williams, Joshua Murry; Landsberger, Sheldon; Pryor, Mitchell WayneThe design of manufacturing systems in hazardous environments is complex, requiring interdisciplinary knowledge to determine which components and operators (human or robotic) are feasible. When conceptualizing designs, some options may be overlooked or unknowingly infeasible due to the design engineers' lack of knowledge in a particular field or ineffective communication of requirements between disciplines. To alleviate many of these design issues, we develop a computational design tool to automate the synthesis of conceptual manufacturing system designs and optimization of preliminary layouts. To generate workcell concepts for manufacturing processes, we create a knowledge-based system (KBS) that performs functional modeling using a common language, a generic component database, and a rule set. The KBS produces high-level task plans for specific manufacturing processes and allocates needed material handling tasks between compatible human and/or robotic labor. We develop an extended pattern search (EPS) algorithm to optimize system layouts based on worker dose and cycle time minimization using the functions and sequencing of generated task plans. The KBS and EPS algorithm were applied to the design of glovebox processing systems at Los Alamos National Laboratory (LANL). Our computational design tool successfully generates design concepts with varied task allocation and processing sub-tasks and layouts with favorable manipulation workspaces. This work establishes a framework for automated conceptual design while providing designers with a beneficial tool for designing manufacturing systems in an interdisciplinary and highly constrained domain.Item Automating X-ray and neutron imaging applications with flexible automation(2015-12) Hashem, Joseph Anthony; Landsberger, Sheldon; Pryor, Mitchell Wayne; Biegalski, Steven; Schneider, Erich; Janecky, DavidThis dissertation advances the capability of autonomous manipulation systems for non-destructive testing applications, specifically computed tomography and radiography. Non-destructive testing is the inspection of a part that does not affect its future usefulness. Radiography and tomography technologies are used to detect material faults inaccessible to direct observation. An industrial 7 degree-of-freedom manipulator has been installed in various x-ray and neutron imaging facilities, including the Nuclear Engineering Teaching Laboratory and Los Alamos National Laboratory, for imaging purposes. Inspection of numerous components manually is laborious and time consuming, and there is the risk of high radiation dose to the operator. As Low As Reasonably Achievable exposure can be significantly reduced by installing a robot in an x-ray or neutron imaging facility to perform part placement in the beam for radioactive parts and nuclear facilities. Automation has the additional potential benefit of improving part throughput by obviating the need for human personnel to move or exchange parts to be imaged and allowing for flexible orientation of the imaged object with respect to the x-ray or neutron beam. When the process is fully automated, it eliminates the need for a human to enter the beam area. The robot needs to meet certain performance requirements, including high repeatability, precision, stability, and accuracy. The robotic system must be able to precisely position and align parts, and parts need to be held still while the image is taken. Any movement of the specimen during exposure causes image blurring. Robotics and remote systems are an integral part of the ALARA approach to radiation safety. Robots increase the distance between workers and hazards and reduce time that workers must be exposed. Research performed aims to expand the role of automation at nuclear facilities by reducing the burden on human operators. The robot’s control system must manage collision detection, grasping, and motion planning to reduce the amount of time that an operator spends micro-managing such a system via tele-operation. The subject of this work includes modeling (in MCNP) and measuring flux, dose rates, and DPA rates of neutron imaging facilities to develop predictions of radiation flux, dose profiles, and radiation damage by examining neutron and gamma fields during operation. Dose and flux predictions provide users the means to simulate geometrical and material changes and additions to a facility, thus saving time, money, and energy in determining the optimal setup for the robotic system.Item Autonomous qualitative learning of distinctions and actions in a developing agent(2010-08) Mugan, Jonathan William; Kuipers, Benjamin; Stone, Peter, 1971-; Ballard, Dana; Cohen, Leslie; Mooney, RaymondHow can an agent bootstrap up from a pixel-level representation to autonomously learn high-level states and actions using only domain general knowledge? This thesis attacks a piece of this problem and assumes that an agent has a set of continuous variables describing the environment and a set of continuous motor primitives, and poses a solution for the problem of how an agent can learn a set of useful states and effective higher-level actions through autonomous experience with the environment. There exist methods for learning models of the environment, and there also exist methods for planning. However, for autonomous learning, these methods have been used almost exclusively in discrete environments. This thesis proposes attacking the problem of learning high-level states and actions in continuous environments by using a qualitative representation to bridge the gap between continuous and discrete variable representations. In this approach, the agent begins with a broad discretization and initially can only tell if the value of each variable is increasing, decreasing, or remaining steady. The agent then simultaneously learns a qualitative representation (discretization) and a set of predictive models of the environment. The agent then converts these models into plans to form actions. The agent then uses those learned actions to explore the environment. The method is evaluated using a simulated robot with realistic physics. The robot is sitting at a table that contains one or two blocks, as well as other distractor objects that are out of reach. The agent autonomously explores the environment without being given a task. After learning, the agent is given various tasks to determine if it learned the necessary states and actions to complete them. The results show that the agent was able to use this method to autonomously learn to perform the tasks.Item Characterizing argumentation structure within the asynchronous, online communication of novice engineering design students(2014-12) McKenna, William F., active 21st century; Treisman, UriPracticing argumentation in secondary school classrooms benefits students both in terms of learning how to argue and learning the course material at hand. Amidst the onset and growth of engineering design courses in secondary schools, this dissertation is an exploratory case study to characterize the use of argumentation among novice student engineering designers. The setting is a high school robotics class. Specifically, a group of students from one class section teamed up with a group of students from a separate class section to design and build a single robot. The team members communicated online via a shared, editable document. That text is the primary data set for my analysis. I looked for indications of argumentation structure that emerged from the online discussion, given that, to my knowledge, the students had not been taught argumentation strategies, per se. Engineering design is relatively new to secondary school, so I thought it appropriate to develop a baseline—a case study that reveals how students communicate about their designs when left largely to their own devices. This study may inform the development argumentation scaffolds that support the students’ existing strengths while ameliorating their weaknesses. My analytical supposition was that argumentation in design will take the form of resolving differences of opinion toward the creation of a single design. Hence, I used Pragma-dialectic theory as my analytical framework. It is a broad theory, based upon resolving differences of opinion in everyday conversation. As such, Pragma-dialectic theory may also be able to encompass the idiosyncrasies of team design, such as reliance on intuition and experience, as well as the important roles that designed objects play throughout the process. Taken together, the importance of intuition, experience, and objects suggests multiple modes of communication that ought to be considered arguments within design deliberations. Results suggest that the students worked to resolve differences of design opinions. In doing so, the students relied heavily on their designed objects to make their arguments meaningful. I classified five object-based claims which emerged from the students’ discussions: keystone, tinkering, visual, tactile, and counterfactual. These form the beginnings of a theory of object-based argumentation.Item Criteria based evaluation of stopping trajectories in serial manipulators(2009-12) Steinfeld, Bryan Christopher; Tesar, Delbert; Pryor, MitchIn the past few years, there has been a large push towards adapting traditional industrial manipulators to other, more consumer-centric applications [1]. These include not only house and elderly care, but also towards medical applications that manipulators may be especially suited for, such as rehabilitation of patients who have suffered neurological trauma [2]. Impeding this push are the strict safety requirements necessary to certify a manipulator for use. These requirements include low speed operation and preventing humans from entering the manipulator workspace [3]. These restrictions effectively prevent a manipulator from being used in many of these applications. Previous work done in manipulator safety research has focused on improving the system’s knowledge of its environment and controlling the manipulator’s motion to keep away from potential hazards. These methods are extremely important in terms of avoiding potential collisions but provide little insight into the situation that occurs once a hazard occurs and the manipulator is forced to react. In order to improve upon the ability to evaluate a manipulator’s overall safety, this report establishes a framework to evaluate the capacity of a manipulator to safely “halt” itself. Two sets of criteria are presented in this report. The first set seeks to quantify both the potential of the manipulator to avoid a collision during the stopping motion and the potential severity of the collision. The second set of criteria quantifies the effect of the stopping motion at the actuator level, allowing the operator to identify potential hardware faults and the capacity to which the actuators are performing. A framework for mapping the manipulator’s actuator parameters for the gear reduction ratio and the motor torque to the potential safety criteria performance is formulated to allow the manipulator designer to match task requirements to the manipulator design. Finally, an examination of the effects on operating parameters such as manipulator configuration, end-effector load, and operating speed is presented with a 6DOF industrial manipulator. This analysis showed that the operating speed of the manipulator is the most important determinant of the safety performance, with the distance traveled by the manipulator increasing by a factor of 15 for all configurations when the speed is increased only by a factor of four. Recommendations for the application of these criteria are presented to the reader as well.Item Design and analysis of a coding and classification system for a systematic interactive computer-aided robot selection procedure (CARSP)(Texas Tech University, 1984-08) Offodile, Onyebuchi FelixA coding and classification system (ROBOCODE) was developed for robots, and used to model a computer based robot selection algorithm. Forty attributes were used to design a taxonomic system for robots and provided a fast and easy standard basis for comparing robots. The system is semi-polycode structure and was readily computerized for easy storage and retrieval of information on robots. The goal of the ROBOCODE system was a user oriented computer-aided robot selection procedure (CARSP). The CARS system software is interactive, and its design showed that the ease with which the vast amount of data on robots, and the number of robots, could be handled was limited primarily by the disc storage space of the computer rather than the computer memory. Coding and classification was found to augment this storage space by a factor of about ten. The same coding system was used to code the tasks the robot was to perform in order to establish an effective matching procedure between the task and the robot. For some task variables that could not be matched directly, a indirect matching procedure was developed. A set of cost equations was developed and used to measure the performance of the robots under shop conditions The necessary condition for selecting a robot was that the codes for the robot be as good as or superior to the corresponding task codes, and the sufficient condition was that the robot had a total minimum operating cost. The robot selection model was evaluated using a statistical procedure to investigate the stability of the model selecting the cost effective robot. Experimental result; showed that the model was fairly stable in selecting this cost effective robot based on the robots' first period total operating cost. For a robot selection problem in which one machine was used to perform one type of task on one type of product (1/1/1), the cost effective robot was selected 67' of the time, and 54% of the time for the robot selection problem in which one machine was used to perform one type of task on four different types of products (1/1/4).Item Development of a design methodology and application to advance the field of highly mobile robotics(2011-05) Pace, Patrick Wayne; Wood, Kristin L.; Wood, John J.Developing innovative ideas as part of engineering design can be limited by the field of technology and the engineer's or design team's understanding of the field. Without sufficient understanding of an emerging technical field, ideation may be hampered by reinventing the proverbial wheel or by a lack of knowledge of the underlying physical principles and state of technology. The research presented here seeks to develop a tool and methodology intended to strengthen a designer’s or design team’s understanding of a field and relevant technologies in order to foster creative and innovative solutions. The presented inductive methodology consists of conducting a thorough review of existing relevant developing or commercially available technologies in order to obtain characteristic property data to be used as a basis of understanding. Analysis of the plotted data may lead to understanding existing trends, identifying voids where opportunities exist to expand the design space and general insights into the field. The effectiveness of using empirical data to look for innovation is investigated in the domain of highly mobile robots. Senior cadets from USAFA and UT Austin perform concept generation sessions before and after utilizing the proposed methodology to validate the effectiveness of the approach. The study at UT Austin validates the proposed methodology by measuring the quantity, quality, and novelty of the concepts generated before and after exposure to the methodology. These experiments demonstrate that state-of-technology design tools provide an effective foundation and platform for designers to generate a larger quantity of concepts. To further investigate the effectiveness of the proposed methodology, it is used to develop a device within the field of highly mobile robotics. There exist applications of highly mobile robots which require innovative solutions with regard to overcoming obstacles, payload capacity, energy storage and minimizing power requirements. The methodology allows for the development of innovative concepts, and the embodiment and manufacture of a particular solution. The mechanical design solutions to multiple design challenges are presented, and the prototyped device proves capable of expanding the existing design space in terms of its performance with respect to the metrics mentioned above.Item Development of mobile platform for inventory and inspection applications in nuclear environments(2015-12) Anderson, Robert Blake; Landsberger, Sheldon; Pryor, Mitchell WayneThe efforts made towards deploying a mobile robotic system at Los Alamos National Laboratory are detailed in this thesis. The platform application is non-contact tasks related to inspection, inventory, and radiation surveying. It is intended for a Special Nuclear Material storage facility featuring a high radiation environment and a variety of storage modes. New robotic capabilities have been developed using several mobile platforms to address the requirements of this application. Many of challenges are common to any warehouse application, such as autonomous task planning, vision, navigation, and inventory data management. Others are specific to a nuclear laboratory environment, such as radiation measurement and analysis, response to radioactive contamination, criticality safety, and restrictive security measures. This thesis describes the progress made towards meeting these challenges, outstanding issues, and future work that is necessary to complete the project. Nuclear facilities are under ever-increasing demands to reduce worker radiation exposure. Since the vault is a high radiation area, it is one of the first targets at Los Alamos for the application of novel solutions. The deployment of this system promises to enhance worker safety by reducing their presence inside the vault and therefore total occupational dose. As robotic systems become more trusted in the nuclear weapons complex, it also has the potential to reduce total operator labor by performing time-consuming tasks autonomously.Item Efficient muscle representation for human walking(2012-12) Iyer, Rahul R.; Ballard, Dana Harry; Fussell, Don; Miikkulainen, Risto; Sentis, Luis; Stone, PeterResearch in robotics has recently broadened its traditional focus on industrial applications to include natural, human-like systems. The human musculoskeletal system has over 600 muscles and 200 joint degrees-of-freedom that provide extraordinary flexibility in tailoring its overall configuration and dynamics to the demands of different tasks. The importance of understanding human movement has spurred efforts to build systems with similar capabilities and has led to the construction of actuators, such as pneumatic artificial muscles, that have properties similar to those of human muscles. However, muscles are far more complex than these robotic actuators and will require new control perspectives. Specifying how to encode high degree-of-freedom muscle functions in order to recreate such movements in anthropomorphic robotic systems is an imposing challenge. This dissertation attempts to advance our understanding by modeling the workings of human muscles in a way that explains how the low temporal bandwidth control of the human brain could direct the high temporal bandwidth requirements of the human movement system. We extend the motor primitives model, a popular strategy for human motor control, by coding a fixed library of movements such that their temporal codes are pre-computed and can be looked up and combined on demand. In this dissertation we develop primitives that lead to various smooth, natural human movements and obtain a sparse-code representation for muscle fiber length changes by applying Matching Pursuit on a parameterized representation of such movements. We employ accurate three-dimensional musculoskeletal models to simulate the lower body muscle fiber length changes for multiple repeatable movements captured from human subjects. We recreate the length changes and show that the signal can be economically encoded in terms of discrete movement elements. Each movement can thus be visualized as a sequence of coefficients for temporally displaced motor primitives. The primary research contribution of describing movements as a compact code develops a clear hierarchy between the spinal cord and higher brain areas. The code has several other advantages. First, it provides an overview of how the elaborate computations in abstract motor control could be ‘parcellated’ into the brain’s primary subsystems. Second, its parametric description could be used in the extension of learned movements to similar movements with different goals. Thirdly, the sensitivity of the parameters can allow the differentiation of very subtle variations in movement. This research lays the groundwork for understanding and developing further human motor control strategies and provides a mathematical framework for experimental research.Item Exploration of an unknown space by collective robotics using fuzzy logic and reinforcement learning(Texas Tech University, 2000-05) Pandya, Ashish K.This thesis concerns itself with the specific problem as follows: search an area using mobile robots without the aid of human (or central) tele-operation. The robots must correctly identify the goal source which is characterized by a maximum intensity (or favorability.) Subsequently, they must reach the position of the goal source while incurring a low total cost (energy consumed). The principles with their scalability and usability are used as evaluation criteria for the methods used to explore the unknown search area. Two different approaches are considered to solve this problem. The first, uses fiizzy mles [1], so that a robot in collaboration with other robots may use the knowledge of its present state vectors to find the desired signal source.The second approach uses reinforced leaming technique to train robots. In this technique, we have 3 different methodologies. The first is the simplest reinforcement leaming called QLeaming in which we have a lookup table to train individual robot. Second method is similar to Simple ACD viz: Heuristic Dynamic Programming (HDP) called Temporal Difference (TD(X)) method. The Temporal Difference method is an elegant way of doing reinforcement leaming. A simple ACD uses two neural networks, e.g., a criticand an action (control) network. The critic network leams to predict the total fiiture cost from a given environment to the terminal state, while the action network leams a policy function to optimize critic's cost output at each state. A graphical user interface and display plus a software implemented simulator are used for experimental purposes for both approaches.Item From the mind to the hand : a beginners guide to milling(2013-08) Crocker, Paul Michael; Deshpande, Ashish D.; Marshall, JillThe purpose of this document is to provide a starting point to engage high school students in a program utilizing CNC milling technology along with CAD/CAM innovations. The beginners’ user manual gives students foundational knowledge in using for Autodesk Inventor, sprutCAM software, and operation of a Tormach PCNC 1100 mill. These tools were chosen since they are available to the author’s school district. Supporting information is given to support the claim of its importance in the classroom for high schoolers.Item GCCF : a generalized contact control framework(2016-05) Von Sternberg, Rusty Alexander; Landsberger, Sheldon; Pryor, Mitchell WayneThe field of robotics has come a long way since the first reprogrammable robot was able to automate simple tasks on an assembly line. However, many industrial robots are stuck doing similar simple tasks in the field, especially in the nuclear industry. Roboticists can expand the task space of industrial robots by making advanced robot technology reliable, easily integrated, and packaged in a manner that does not require an expert in the field to use. One particular field of robotics that could be used to help this task space expansion is compliant control which is used to execute robotic procedures involving contact with environmental objects. It is especially useful when the position or orientation of the environmental objects is not precise. Examples of industrial procedures that a robot could do with compliant control include material reduction, surface finishing, packaging, assembly, material handling, and many more. This thesis explores the state of the art in compliant control and proposes a Generalized Contact Control Framework (GCCF) that packages compliant control laws in a manner that is easy to use for a non-expert. GCCF splits the control of a robot end effector into separate control of each linear and rotational dimension. The user sets the law that controls each dimension independently to one of three intuitive laws. By specifying laws and stiffness independently for each dimension of end effector control, the user can complete a large variety of contact tasks. We illustrate GCCF’s broad capabilities in two flexible demonstrations. The first demonstration provides a graphical user interface to GCCF with which a user can set and reconfigure the control of the end effector while interacting with the robot. This allows the user to subjectively experience the reconfigurablilty as well as the physical behavior prompted by the control. In the second demonstration, we use GCCF to execute multiple contact tasks with the goal of putting a peg in a hole. These demonstrations prove the feasibility and usefulness of GCCF, using the API and ROS compatible package for the controller.Item Geometric workcell modeling for robot control and coordination(2007-05) Knoll, Jonathan Andrew, 1980-; Tesar, DelbertThe expansion of robotics to new industries and advances in technology brings them into closer proximity to humans which necessitates careful robot motion in relation to the environment. This report addresses the need for better geometric modeling of robotic workcells for control and coordination in terms motion planning, obstacle avoidance, and collision detection. With the realization that input modeling parameters greatly influence the type and quality of the resultant output state of the model, input/output parameters are closely examined in this report. This examination puts particular emphasis on the relationship between geometric representation and separation distance. Modeling techniques are then studied and compared to determine one which will accomplish a set of requirements for a complex robotic workcell. The Trauma Pod is a remotely controlled robotic operating room where the patient is the only human in the room. In this report, the Trauma Pod served as a case study for applying modeling techniques to fulfill requirements though a modeling framework. The requirements for the Trauma Pod included supporting collision detection and distance calculation between high resolution geometry and providing other custom features such as calculation of manipulator self-collisions. A general set of requirements were then established, and similar capabilities developed for the Trauma Pod were extended to general robotic workcells for satisfaction of those requirements. This established a set of modeling tools for a geometric workcell modeling framework. Accomplishment of all Trauma Pod modeling requirements proved the validity of the modeling technique chosen in this report. Furthermore, the chosen modeling technique has been implemented into Operational Software Components for Advanced Robotics (OSCAR) to provide powerful modeling capabilities to general robotic workcells. This work has not only proven to dramatically improve the collision detection capability of OSCAR, but it also provides potential benefit to robot motion planning and obstacle avoidance.Item Graph-based world-model for robotic manipulation(2010-08) O'Neil, Brian Erick, 1978-; Landsberger, Sheldon; Pryor, Mitchell WayneThere has been a significant push in robotics research toward robot autonomy. However, full autonomy is currently impractical for all but the most clearly defined tasks in the most structured environments. However, as tasks become less defined and environments become cluttered and less controlled, there is still a benefit to implementing semi-autonomous behaviors where aspects of the tasks are completed autonomously thus reducing the burden on the human operator. A key component of a robot control system that supports this functionality is a robust world model to act as a repository of environmental information. The research community has provided many world-modeling solutions to support autonomous vehicle navigation. As such, they focus primarily on preventing collisions with the environment. Modeling schemes designed for collision prevention are of limited use to robotic manipulators that must have contact interaction with the environment as a matter of course. This thesis presents a world-modeling scheme that abstracts the model of the environment into a graph structure. This abstraction separates the concepts of entities in the environment from their relationships to the environment. The result is an intuitive world model that supports not only collision detection, but also motion planning and grasping. The graph-based world model presented can be searched by semantic type and tag values, allowing any number of agents to simultaneously use and update the model without causing failures elsewhere in the system. These capabilities are demonstrated on two different automated hot-cell glovebox systems, and one mobile manipulation system for use in remote contamination testing.Item High-performance series elastic actuation(2014-08) Paine, Nicholas Arden; Sentis, Luis; Vishwanath, SriramMobile legged robots have the potential to restructure many aspects of our lives in the near future. Whether for applications in household care, entertainment, or disaster response, these systems depend on high-performance actuators to improve their basic capabilities. The work presented here focuses on developing new high-performance actuators, specifically series elastic actuators, to address this need. We adopt a system-wide optimization approach, dealing with factors which influence performance at the levels of mechanical design, electrical system design, and control. Using this approach and based on a set of performance metrics, we produce an actuator, the UT-SEA, which achieves leading empirical results in terms of power-to-weight, force control, size, and system efficiency. We also develop general high-performance control techniques for both force- and position-controlled actuators, some of which were adopted for use on NASA-JSC's Valkyrie Humanoid robot and were used during DARPA's DRC Trials 2013 robotics competition.Item Improved manipulator configurations for grasping and task completion based on manipulability(2010-12) Williams, Joshua Murry; Pryor, Mitchell Wayne; Landsberger, SheldonWhen a robotic system executes a task, there are a number of responsibilities that belong to either the operator and/or the robot. A more autonomous system has more responsibilities in the completion of a task and must possess the decision making skills necessary to adequately deal with these responsibilities. The system must also handle environmental constraints that limit the region of operability and complicate the execution of tasks. There are decisions about the robot’s internal configuration and how the manipulator should move through space, avoid obstacles, and grasp objects. These motions usually have limits and performance requirements associated with them. Successful completion of tasks in a given environment is aided by knowledge of the robot’s capabilities in its workspace. This not only indicates if a task is possible but can suggest how a task should be completed. In this work, we develop a grasping strategy for selecting and attaining grasp configurations for flexible tasks in environments containing obstacles. This is done by sampling for valid grasping configurations at locations throughout the workspace to generate a task plane. Locations in the task plane that contain more valid configurations are stipulated to have higher dexterity and thus provide greater manipulability of targets. For valid configurations found in the plane, we develop a strategy for selecting which configurations to choose when grasping and/or placing an object at a given location in the workspace. These workspace task planes can also be utilized as a design tool to configure the system around the manipulator’s capabilities. We determine the quality of manipulator positioning in the workspace based on manipulability and locate the best location of targets for manipulation. The knowledge of valid manipulator configurations throughout the workspace can be used to extend the application of task planes to motion planning between grasping configurations. This guides the end-effector through more dexterous workspace regions and to configurations that move the arm away from obstacles. The task plane technique employed here accurately captures a manipulator’s capabilities. Initial tests for exploiting these capabilities for system design and operation were successful, thus demonstrating this method as a viable starting point for incrementally increasing system autonomy.Item Improving the safety and efficiency of rail yard operations using robotics(2013-12) Boddiford, Andrew Shropshire; Tesar, DelbertSignificant efforts have been expended by the railroad industry to make operations safer and more efficient through the intelligent use of sensor data. This work proposes to take the technology one step further to use this data for the control of physical systems designed to automate hazardous railroad operations, particularly those that require humans to interact with moving trains. To accomplish this, application specific requirements must be established to design self-contained machine vision and robotic solutions to eliminate the risks associated with existing manual operations. Present-day rail yard operations have been identified as good candidates to begin development. Manual uncoupling, in particular, of rolling stock in classification yards has been investigated. To automate this process, an intelligent robotic system must be able to detect, track, approach, contact, and manipulate constrained objects on equipment in motion. This work presents multiple prototypes capable of autonomously uncoupling full-scale freight cars using feedback from its surrounding environment. Geometric image processing algorithms and machine learning techniques were implemented to accurately identify cylindrical objects in point clouds generated in real-vi time. Unique methods fusing velocity and vision data were developed to synchronize a pair of moving rigid bodies in real-time. Multiple custom end-effectors with in-built compliance and fault tolerance were designed, fabricated, and tested for grasping and manipulating cylindrical objects. Finally, an event-driven robotic control application was developed to safely and reliably uncouple freight cars using data from 3D cameras, velocity sensors, force/torque transducers, and intelligent end-effector tooling. Experimental results in a lab setting confirm that modern robotic and sensing hardware can be used to reliably separate pairs of rolling stock up to two miles per hour. Additionally, subcomponents of the autonomous pin-pulling system (APPS) were designed to be modular to the point where they could be used to automate other hazardous, labor-intensive tasks found in U.S. classification yards. Overall, this work supports the deployment of autonomous robotic systems in semi-unstructured yard environments to increase the safety and efficiency of rail operations.
- «
- 1 (current)
- 2
- 3
- »