Matching Items (41)
Filtering by
- All Subjects: robotics
- Creators: Yang, Yezhou
- Creators: Redkar, Sangram
- Resource Type: Text
Description
Robotic rehabilitation for upper limb post-stroke recovery is a developing technology. However, there are major issues in the implementation of this type of rehabilitation, issues which decrease efficacy. Two of the major solutions currently being explored to the upper limb post-stroke rehabilitation problem are the use of socially assistive rehabilitative robots, robots which directly interact with patients, and the use of exoskeleton-based systems of rehabilitation. While there is great promise in both of these techniques, they currently lack sufficient efficacy to objectively justify their costs. The overall efficacy to both of these techniques is about the same as conventional therapy, yet each has higher overhead costs that conventional therapy does. However there are associated long-term cost savings in each case, meaning that the actual current viability of either of these techniques is somewhat nebulous. In both cases, the problems which decrease technique viability are largely related to joint action, the interaction between robot and human in completing specific tasks, and issues in robot adaptability that make joint action difficult. As such, the largest part of current research into rehabilitative robotics aims to make robots behave in more "human-like" manners or to bypass the joint action problem entirely.
ContributorsRamakrishna, Vijay Kambhampati (Author) / Helms Tillery, Stephen (Thesis director) / Buneo, Christopher (Committee member) / Barrett, The Honors College (Contributor) / Economics Program in CLAS (Contributor) / W. P. Carey School of Business (Contributor) / School of Life Sciences (Contributor)
Created2015-05
Description
Reinforcement Learning(RL) algorithms have made a remarkable contribution in the eld of robotics and training human-like agents. On the other hand, Evolutionary Algorithms(EA) are not well explored and promoted to use in the robotics field. However, they have an excellent potential to perform well. In thesis work, various RL learning algorithms like Q-learning, Deep Deterministic Policy Gradient(DDPG), and Evolutionary Algorithms(EA) like Harmony Search Algorithm(HSA) are tested for a customized Penalty Kick Robot environment. The experiments are done with both discrete and continuous action space for a penalty kick agent. The main goal is to identify which algorithm suites best in which scenario. Furthermore, a goalkeeper agent is also introduced to block the ball from reaching the goal post using the multiagent learning algorithm.
ContributorsTrivedi, Maitry Ronakbhai (Author) / Amor, Heni Ben (Thesis advisor) / Redkar, Sangram (Thesis advisor) / Sugar, Thomas (Committee member) / Arizona State University (Publisher)
Created2021
Description
Traditional Reinforcement Learning (RL) assumes to learn policies with respect to reward available from the environment but sometimes learning in a complex domain requires wisdom which comes from a wide range of experience. In behavior based robotics, it is observed that a complex behavior can be described by a combination of simpler behaviors. It is tempting to apply similar idea such that simpler behaviors can be combined in a meaningful way to tailor the complex combination. Such an approach would enable faster learning and modular design of behaviors. Complex behaviors can be combined with other behaviors to create even more advanced behaviors resulting in a rich set of possibilities. Similar to RL, combined behavior can keep evolving by interacting with the environment. The requirement of this method is to specify a reasonable set of simple behaviors. In this research, I present an algorithm that aims at combining behavior such that the resulting behavior has characteristics of each individual behavior. This approach has been inspired by behavior based robotics, such as the subsumption architecture and motor schema-based design. The combination algorithm outputs n weights to combine behaviors linearly. The weights are state dependent and change dynamically at every step in an episode. This idea is tested on discrete and continuous environments like OpenAI’s “Lunar Lander” and “Biped Walker”. Results are compared with related domains like Multi-objective RL, Hierarchical RL, Transfer learning, and basic RL. It is observed that the combination of behaviors is a novel way of learning which helps the agent achieve required characteristics. A combination is learned for a given state and so the agent is able to learn faster in an efficient manner compared to other similar approaches. Agent beautifully demonstrates characteristics of multiple behaviors which helps the agent to learn and adapt to the environment. Future directions are also suggested as possible extensions to this research.
ContributorsVora, Kevin Jatin (Author) / Zhang, Yu (Thesis advisor) / Yang, Yezhou (Committee member) / Praharaj, Sarbeswar (Committee member) / Arizona State University (Publisher)
Created2021
Description
The increasing availability of data and advances in computation have spurred the development of data-driven approaches for modeling complex dynamical systems. These approaches are based on the idea that the underlying structure of a complex system can be discovered from data using mathematical and computational techniques. They also show promise for addressing the challenges of modeling high-dimensional, nonlinear systems with limited data. In this research expository, the state of the art in data-driven approaches for modeling complex dynamical systems is surveyed in a systemic way. First the general formulation of data-driven modeling of dynamical systems is discussed. Then several representative methods in feature engineering and system identification/prediction are reviewed, including recent advances and key challenges.
ContributorsShi, Wenlong (Author) / Ren, Yi (Thesis advisor) / Hong, Qijun (Committee member) / Jiao, Yang (Committee member) / Yang, Yezhou (Committee member) / Arizona State University (Publisher)
Created2022
Description
This research proposes some new data-driven control methods to control a nonlinear dynamic model. The nonlinear dynamic model linearizes by using the Koopman theory. The Koopman operator is the most important part of designing the Koopman theory. The data mode decomposition (DMD) is used to obtain the Koopman operator. The proposed data-driven control method applies to different nonlinear systems such as microelectromechanical systems (MEMS), Worm robots, and 2 degrees of freedom (2 DoF) robot manipulators to verify the performance of the proposed method. For the MEMS gyroscope, three control methods are applied to the linearized dynamic model by the Koopman theory: linear quadratic regulator (LQR), compound fractional PID sliding mode control, and fractional order PID controller tuned with bat algorithm. For the Worm robot, an LQR controller is proposed to control the linearized dynamic model by the Koopman theory. A new fractional sliding mode control is proposed to control the 2 DoF arm robot. All the proposed controllers applied to the linearized dynamic model by the Kooman theory are compared with some conventional proposed controllers such as PID, sliding mode control, and conventional fractional sliding mode control to verify the performance of the proposed controllers. Simulation results validate their performance in high tracking performance, low tracking error, low frequency, and low maximum overshoot.
ContributorsRahmani, Mehran (Author) / Redkar, Sangram (Thesis advisor) / Sugar, Thomas (Committee member) / C. Subramanian, Susheelkumar (Committee member) / Arizona State University (Publisher)
Created2023
Description
Multiple robotic arms collaboration is to control multiple robotic arms to collaborate with each other to work on the same task. During the collaboration, theagent is required to avoid all possible collisions between each part of the robotic arms.
Thus, incentivizing collaboration and preventing collisions are the two principles which
are followed by the agent during the training process. Nowadays, more and more
applications, both in industry and daily lives, require at least two arms, instead of
requiring only a single arm. A dual-arm robot satisfies much more needs of different
types of tasks, such as folding clothes at home, making a hamburger in a grill or
picking and placing a product in a warehouse.
The applications done in this paper are all about object pushing. This thesis
focuses on how to train the agent to learn pushing an object away as far as possible.
Reinforcement Learning (RL), which is a type of Machine Learning (ML), is then
utilized in this paper to train the agent to generate optimal actions. Deep Deterministic
Policy Gradient (DDPG) and Hindsight Experience Replay (HER) are the two RL
methods used in this thesis.
ContributorsLin, Steve (Author) / Ben Amor, Hani (Thesis advisor) / Redkar, Sangram (Committee member) / Zhang, Yu (Committee member) / Arizona State University (Publisher)
Created2023
Description
The need for robust verification and validation of automated vehicles (AVs) to ensure driving safety grows more urgent as increasing numbers of AVs are allowed to operate on open roads. To address this need, AV developers can present a safety case to regulators and the public that provides an evidence-based justification of their assertion that an AV is safe to operate on open roads. This work aims to describe the development of a scenario-based testing methodology that contributes to this safety case. A high-level definition of this test selection and scoring methodology (TSSM) is first presented, along with an outline of its scope and key ideas. This is followed by a literature review that details the current state of the art in AV testing, including the driving performance metrics and equations that provide a basis for the TSSM. A chart-based method for quantifying an AV’s operational design domain (ODD) and behavioral competency portfolio is then described that provides the foundation for a scenario generation and filtration process. After outlining a method for the AV to progress through increasingly robust test methods based on its current technology readiness level (TRL), the generation and filtration of two sets of scenarios by the TSSM is outlined: a standardized set that can be used to compare the performance of vehicles with identical ODD and behavioral competency portfolios, and a set containing high-relevance scenarios that is partially randomized to ensure test integrity. A related framework for incorporating testing on open roads is subsequently specified. An equation for an overall AV driving performance score is then defined that quantifies the aggregate performance of the AV across all generated scenarios. The TSSM continues according to an iterative process, which includes a method for exploring edge and corner scenarios, until a stopping condition is achieved. Two proofs of concept are provided: a demonstration of the ability of the TSSM to pare scenarios from a preexisting database, and an example ODD and behavioral competency portfolio specification form. Finally, this work concludes by evaluating the TSSM and its proofs of concept and outlining possible future work on the methodology.
ContributorsO'Malley, Gavin (Author) / Wishart, Jeffrey (Thesis advisor) / Zhao, Junfeng (Thesis advisor) / Yang, Yezhou (Committee member) / Arizona State University (Publisher)
Created2023
Description
This work endeavors to lay a solid foundation for the exploration and the considerations of exoskeletons, exosuits, and medical devices concerning proprioceptive feedback. This investigation is situated at the nexus of engineering, neuroscience, and rehabilitation medicine, striving to cultivate a holistic understanding of how mechanical augmentation, interfaced synergistically with human proprioception, can foster enhanced mobility and safety. This is especially pertinent for individuals with compromised motor functions.British Neurologist Oliver Wolf Sacks in 1985 published “The Man who Mistook His Wife for a Hat” a series of his most memorable neurological case describing the brain's strangest pathways. One of these cases is “The Disembodied Lady”, Christina a 27-year-old woman that lost entirely the sense of proprioception due to polyneuropathy. This caused her to not be able to control her body, and she declares that “I feel the wind on my arms and face, and then I know, faintly, I have arms and a face. It’s not the real thing, but it’s something—it lifts this horrible, dead veil for a while. ” Finally, she was able to control her body using vision alone. Dr. Sacks introduced, for the first time, the importance of proprioception, as the sense of position of body parts relative to other parts of the body, to western culture. This document’s mission is to identify unexplored concepts in the literature regarding exoskeletons, wearables and assistive technology and a user’s proprioception, embodiment and utilization when wearing devices. Dr. Philipp Beckerle suggests the need to research the connections between wearable hardware and human sense of proprioception. He also emphasizes the need for functional assessment protocols for wearables devices and the role of embodiment. He criticizes the current commercially available upper-limb prostheses since they only restore limited functions and therefore impede embodiment.
This document’s goal is to identify operative solutions through the adaptation of existing technologies and to use effective solutions to improve the quality of life of people suffering from pathologies or traumatic injuries.
ContributorsVignola, Claudio (Author) / Sugar, Thomas (Thesis advisor) / Redkar, Sangram (Committee member) / McDaniels, Troy (Committee member) / Arizona State University (Publisher)
Created2023
Description
In recent years, there has been a growing emphasis on developing automated systems to enhance traffic safety, particularly in the detection of dilemma zones (DZ) at intersections. This study focuses on the automated detection of DZs at roundabouts using trajectory forecasting, presenting an advanced system with perception capabilities. The system utilizes a modular, graph-structured recurrent model that predicts the trajectories of various agents, accounting for agent dynamics and incorporating heterogeneous data such as semantic maps. This enables the system to facilitate traffic management decision-making and improve overall intersection safety. To assess the system's performance, a real-world dataset of traffic roundabout intersections was employed. The experimental results demonstrate that our Superpowered Trajectron++ system exhibits high accuracy in detecting DZ events, with a false positive rate of approximately 10%. Furthermore, the system has the remarkable ability to anticipate and identify dilemma events before they occur, enabling it to provide timely instructions to vehicles. These instructions serve as guidance, determining whether vehicles should come to a halt or continue moving through the intersection, thereby enhancing safety and minimizing potential conflicts. In summary, the development of automated systems for detecting DZs represents an important advancement in traffic safety. The Superpowered Trajectron++ system, with its trajectory forecasting capabilities and incorporation of diverse data sources, showcases improved accuracy in identifying DZ events and can effectively guide vehicles in making informed decisions at roundabout intersections.
ContributorsChelenahalli Satish, Manthan (Author) / Yang, Yezhou (Thesis advisor) / Lu, Duo (Committee member) / Farhadi, Mohammad (Committee member) / Arizona State University (Publisher)
Created2023
Description
In today’s modern world, industrial robots are utilized in hazardous working condi-tions across all industries, including the renewable energy industry. Robot control
systems and sensors receive and transmit information and data obtained from the
users. Over the last ten years, unmanned vehicles have developed into a subject of
interest for a variety of research institutions. Technology breakthroughs are redefin-
ing disaster relief, search-and-rescue(SAR) and salvage operations’ for aerial robotic
systems as well as terrestrial and marine ones. A team of collaborative robots is
required for the challenging environments, such as space construction, and disaster
relief. These robots will have to make trade-offs between mobility and capabilities
owing to cost, power, and size constraints. Task execution in numerous areas may de-
mand for robot collaboration in order to optimize team performance. An analysis of
collaborative Unmanned Aerial Vehicle(UAV) and Unmanned Ground Vehicle(UGV)
systems is one of the main components of this thesis. UAV/UGV collaborative frame-
works and methods have been presented for reaching or monitoring moving human
targets, a stated set-point for a mobile UGV robot to go to in order to approach
a dynamic target, and actions to take by the UAVs when the mobile UGV robot
is obstructed and cannot reach the target. This method encourages the target and
robot to work together more closely. This is one of the most difficult issues in search
and rescue operations since human targets are seldom found using just land robots or
aerial robots. Finally, the purpose of this thesis is to suggest that the evaluation of
the performance of a collaborative robot system may be accomplished by measuring
the mobility of robots. Even though multi-robot coordination aids in SAR opera-
tions, the findings of the study presented in this thesis conclude that the integration
of various autonomous robotic systems in unstructured environments is difficult and
that there is currently no unitary analytical model that can be used for this purpose.
ContributorsCherupally, SuryaKiran (Author) / Redkar, Sangram (Thesis advisor) / Nichols, Kevin (Committee member) / Subramanian, Susheel Kumar Cherangara (Committee member) / Arizona State University (Publisher)
Created2022