Matching Items (77)
Filtering by
- All Subjects: robotics
- Creators: Zhang, Wenlong
- Creators: Yang, Yezhou
- Creators: Redkar, Sangram
Description
This thesis proposes the concept of soft robotic supernumerary limbs to assist the wearer in the execution of tasks, whether it be to share loads or replace an assistant. These controllable extra arms are made using soft robotics to reduce the weight and cost of the device, and are not limited in size and location to the user's arm as with exoskeletal devices. Soft robotics differ from traditional robotics in that they are made using soft materials such as silicone elastomers rather than hard materials such as metals or plastics. This thesis presents the design, fabrication, and testing of the arm, including the joints and the actuators to move them, as well as the design and fabrication of the human-body interface to unite man and machine. This prototype utilizes two types of pneumatically-driven actuators, pneumatic artificial muscles and fiber-reinforced actuators, to actuate the elbow and shoulder joints, respectively. The robotic limb is mounted at the waist on a backpack frame to avoid interfering with the wearer's biological arm. Through testing and evaluation, this prototype device proves the feasibility of soft supernumerary limbs, and opens up opportunities for further development into the field.
ContributorsOlson, Weston Roscoe (Author) / Polygerinos, Panagiotis (Thesis director) / Zhang, Wenlong (Committee member) / Engineering Programs (Contributor) / Barrett, The Honors College (Contributor)
Created2016-05
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
The mean age of the world’s population is rapidly increasing and with that growth in an aging population a large number of elderly people are in need of walking assistance. In addition, a number of medical conditions contribute to gait disorders that require gait rehabilitation. Wearable robotics can be used to improve functional outcomes in the gait rehabilitation process. The ankle push-off phase of an individual’s gait is vital to their ability to walk and propel themselves forward. During the ankle push-off phase of walking, plantar flexors are required to providing a large amount of force to power the heel off the ground.
The purpose of this project is to improve upon the passive ankle foot orthosis originally designed in the ASU’s Robotics and Intelligent Systems Laboratory (RISE Lab). This device utilizes springs positioned parallel to the user’s Achilles tendon which store energy to be released during the push off phase of the user’s gait cycle. Goals of the project are to improve the speed and reliability of the ratchet and pawl mechanism, design the device to fit a wider range of shoe sizes, and reduce the overall mass and size of the device. The resulting system is semi-passive and only utilizes a single solenoid to unlock the ratcheting mechanism when the spring’s potential force is required. The device created also utilizes constant force springs rather than traditional linear springs which allows for a more predictable level of force. A healthy user tested the device on a treadmill and surface electromyography (sEMG) sensors were placed on the user’s plantar flexor muscles to monitor potential reductions in muscular activity resulting from the assistance provided by the AFO device. The data demonstrates the robotic shoe was able to assist during the heel-off stage and reduced activation in the plantar flexor muscles was evident from the EMG data collected. As this is an ongoing research project, this thesis will also recommend possible design upgrades and changes to be made to the device in the future. These upgrades include utilizing a carbon fiber or lightweight plastic frame such as many of the traditional ankle foot-orthosis sold today and introducing a system to regulate the amount of spring force applied as a function of the force required at specific times of the heel off gait phase.
The purpose of this project is to improve upon the passive ankle foot orthosis originally designed in the ASU’s Robotics and Intelligent Systems Laboratory (RISE Lab). This device utilizes springs positioned parallel to the user’s Achilles tendon which store energy to be released during the push off phase of the user’s gait cycle. Goals of the project are to improve the speed and reliability of the ratchet and pawl mechanism, design the device to fit a wider range of shoe sizes, and reduce the overall mass and size of the device. The resulting system is semi-passive and only utilizes a single solenoid to unlock the ratcheting mechanism when the spring’s potential force is required. The device created also utilizes constant force springs rather than traditional linear springs which allows for a more predictable level of force. A healthy user tested the device on a treadmill and surface electromyography (sEMG) sensors were placed on the user’s plantar flexor muscles to monitor potential reductions in muscular activity resulting from the assistance provided by the AFO device. The data demonstrates the robotic shoe was able to assist during the heel-off stage and reduced activation in the plantar flexor muscles was evident from the EMG data collected. As this is an ongoing research project, this thesis will also recommend possible design upgrades and changes to be made to the device in the future. These upgrades include utilizing a carbon fiber or lightweight plastic frame such as many of the traditional ankle foot-orthosis sold today and introducing a system to regulate the amount of spring force applied as a function of the force required at specific times of the heel off gait phase.
ContributorsSchaller, Marcus Frank (Author) / Zhang, Wenlong (Thesis director) / Sugar, Thomas (Committee member) / Engineering Programs (Contributor) / Barrett, The Honors College (Contributor)
Created2019-12
Description
This work presents the design, modeling, analysis, and experimental characterization and testing of soft wearable robotics for lower limb rehabilitation for the ankle and hip. The Soft Robotic Ankle-Foot Orthosis (SR-AFO) is a wearable soft robot designed using multiple pneumatically-powered soft actuators to assist the ankle in multiple degrees-of-freedom during standing and walking tasks. The flat fabric pneumatic artificial muscle (ff-PAM) contracts upon pressurization and assists ankle plantarflexion in the sagittal plane. The Multi-material Actuator for Variable Stiffness (MAVS) aids in supporting ankle inversion/eversion in the frontal plane. Analytical models of the ff-PAM and MAVS were created to understand how the changing of the design parameters affects tensile force generation and stiffness support, respectively. The models were validated by both finite element analysis and experimental characterization using a universal testing machine. A set of human experiments were performed with healthy participants: 1) to measure lateral ankle support during quiet standing, 2) to determine lateral ankle support during walking over compliant surfaces, and 3) to evaluate plantarflexion assistance at push-off during treadmill walking, and 4) determine if the SR-AFO could be used for gait entrainment. Group results revealed increased ankle stiffness during quiet standing with the MAVS active, reduced ankle deflection while walking over compliant surfaces with the MAVS active, and reduced muscle effort from the SOL and GAS during 40 - 60% of the gait cycle with the dual ff-PAM active. The SR-AFO shows promising results in providing lateral ankle support and plantarflexion assistance with healthy participants, and a drastically increased basin of entrainment, which suggests a capability to help restore the gait of impaired users in future trials. The ff-PAM actuators were used in an X-orientation to assist the hip in flexion and extension. The Soft Robotic Hip Exosuit (SR-HExo) was evaluated using the same set of actuators and trials with healthy participants showed reduction in muscle effort during hip flexion and extension to further enhance the study of soft fabric actuators on human gait assistance.
ContributorsThalman, Carly Megan (Author) / Lee, Hyunglae (Thesis advisor) / Artemiadis, Panagiotis (Thesis advisor) / Sugar, Thomas (Committee member) / Zhang, Wenlong (Committee member) / Arizona State University (Publisher)
Created2021
Description
Physical and structural tree measurements are applied in forestry, precision agriculture and conservation for various reasons. Since measuring tree properties manually is tedious, measurements from only a small subset of trees present in a forest, agricultural land or survey site are often used. Utilizing robotics to autonomously estimate physical tree dimensions would speed up the measurement or data collection process and allow for a much larger set of trees to be used in studies. In turn, this would allow studies to make more generalizable inferences about areas with trees. To this end, this thesis focuses on developing a system that generates a semantic representation of the topology of a tree in real-time. The first part describes a simulation environment and a real-world sensor suite to develop and test the tree mapping pipeline proposed in this thesis. The second part presents details of the proposed tree mapping pipeline. Stage one of the mapping pipeline utilizes a deep learning network to detect woody and cylindrical portions of a tree like trunks and branches based on popular semantic segmentation networks. Stage two of the pipeline proposes an algorithm to separate the detected portions of a tree into individual trunk and branch segments. The third stage implements an optimization algorithm to represent each segment parametrically as a cylinder. The fourth stage formulates a multi-sensor factor graph to incrementally integrate and optimize the semantic tree map while also fusing two forms of odometry. Finally, results from all the stages of the tree mapping pipeline using simulation and real-world data are presented. With these implementations, this thesis provides an end-to-end system to estimate tree topology through semantic representations for forestry and precision agriculture applications.
ContributorsVishwanatha, Rakshith (Author) / Das, Jnaneshwar (Thesis advisor) / Martin, Roberta (Committee member) / Throop, Heather (Committee member) / Zhang, Wenlong (Committee member) / Ehsani, Reza (Committee member) / Arizona State University (Publisher)
Created2022
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