Matching Items (40)
Filtering by
- All Subjects: Mechanical Engineering
- All Subjects: Reinforcement Learning
- Creators: Berman, Spring
Description
Humans have an inherent capability of performing highly dexterous and skillful tasks with their arms, involving maintaining posture, movement and interacting with the environment. The latter requires for them to control the dynamic characteristics of the upper limb musculoskeletal system. Inertia, damping and stiffness, a measure of mechanical impedance, gives a strong representation of these characteristics. Many previous studies have shown that the arm posture is a dominant factor for determining the end point impedance in a horizontal plane (transverse plane). The objective of this thesis is to characterize end point impedance of the human arm in the three dimensional (3D) space. Moreover, it investigates and models the control of the arm impedance due to increasing levels of muscle co-contraction. The characterization is done through experimental trials where human subjects maintained arm posture, while perturbed by a robot arm. Moreover, the subjects were asked to control the level of their arm muscles' co-contraction, using visual feedback of their muscles' activation, in order to investigate the effect of the muscle co-contraction on the arm impedance. The results of this study showed a very interesting, anisotropic increase of the arm stiffness due to muscle co-contraction. This can lead to very useful conclusions about the arm biomechanics as well as many implications for human motor control and more specifically the control of arm impedance through muscle co-contraction. The study finds implications for the EMG-based control of robots that physically interact with humans.
ContributorsPatel, Harshil Naresh (Author) / Artemiadis, Panagiotis (Thesis advisor) / Berman, Spring (Committee member) / Helms Tillery, Stephen (Committee member) / Arizona State University (Publisher)
Created2013
Description
This research report investigates the feasibility of using RFID in Traffic Sign Recognition (TSR) Systems for autonomous vehicles, specifically driver-less cars. Driver-less cars are becoming more prominent in society but must be designed to integrate with the current transportation infrastructure. Current research in TSR systems use image processing as well as LIDAR to identify traffic signs, yet these are highly dependent on lighting conditions, camera quality and sign visibility. The read rates of current TSR systems in literature are approximately 96 percent. The usage of RFID in TSR systems can improve the performance of traditional TSR systems. An RFID TSR was designed for the Autonomous Pheeno Test-bed at the Arizona State University (ASU) Autonomous Collective Systems (ACS) Laboratory. The system was tested with varying parameters to see the effect of the parameters on the read rate. It was found that high reader strength and low tag distance had a maximum read rate of 96.3 percent, which is comparable to existing literature. It was proven that an RFID TSR can perform as well as traditional TSR systems, and has the capacity to improve accuracy when used alongside RGB cameras and LIDAR.
ContributorsMendoza, Madilyn Kido (Author) / Berman, Spring (Thesis director) / Yu, Hongbin (Committee member) / Mechanical and Aerospace Engineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2018-05
Description
With the substantial development of intelligent robots, human-robot interaction (HRI) has become ubiquitous in applications such as collaborative manufacturing, surgical robotic operations, and autonomous driving. In all these applications, a human behavior model, which can provide predictions of human actions, is a helpful reference that helps robots to achieve intelligent interaction with humans. The requirement elicits an essential problem of how to properly model human behavior, especially when individuals are interacting or cooperating with each other. The major objective of this thesis is to utilize the human intention decoding method to help robots enhance their performance while interacting with humans. Preliminary work on integrating human intention estimation with an HRI scenario is shown to demonstrate the benefit. In order to achieve this goal, the research topic is divided into three phases. First, a novel method of an online measure of the human's reliance on the robot, which can be estimated through the intention decoding process from human actions，is described. An experiment that requires human participants to complete an object-moving task with a robot manipulator was conducted under different conditions of distractions. A relationship is discovered between human intention and trust while participants performed a familiar task with no distraction. This finding suggests a relationship between the psychological construct of trust and joint physical coordination, which bridges the human's action to its mental states. Then, a novel human collaborative dynamic model is introduced based on game theory and bounded rationality, which is a novel method to describe human dyadic behavior with the aforementioned theories. The mutual intention decoding process was also considered to inform this model. Through this model, the connection between the mental states of the individuals to their cooperative actions is indicated. A haptic interface is developed with a virtual environment and the experiments are conducted with 30 human subjects. The result suggests the existence of mutual intention decoding during the human dyadic cooperative behaviors. Last, the empirical results show that allowing agents to have empathy in inference, which lets the agents understand that others might have a false understanding of their intentions, can help to achieve correct intention inference. It has been verified that knowledge about vehicle dynamics was also important to correctly infer intentions. A new courteous policy is proposed that bounded the courteous motion using its inferred set of equilibrium motions. A simulation, which is set to reproduce an intersection passing case between an autonomous car and a human driving car, is conducted to demonstrate the benefit of the novel courteous control policy.
ContributorsWang, Yiwei (Author) / Zhang, Wenlong (Thesis advisor) / Berman, Spring (Committee member) / Lee, Hyunglae (Committee member) / Ren, Yi (Committee member) / Yang, Yezhou (Committee member) / Arizona State University (Publisher)
Created2021
Description
A Graph Neural Network (GNN) is a type of neural network architecture that operates on data consisting of objects and their relationships, which are represented by a graph. Within the graph, nodes represent objects and edges represent associations between those objects. The representation of relationships and correlations between data is unique to graph structures. GNNs exploit this feature of graphs by augmenting both forms of data, individual and relational, and have been designed to allow for communication and sharing of data within each neural network layer. These benefits allow each node to have an enriched perspective, or a better understanding, of its neighbouring nodes and its connections to those nodes. The ability of GNNs to efficiently process high-dimensional node data and multi-faceted relationships among nodes gives them advantages over neural network architectures such as Convolutional Neural Networks (CNNs) that do not implicitly handle relational data. These quintessential characteristics of GNN models make them suitable for solving problems in which the correspondences among input data are needed to produce an accurate and precise representation of these data. GNN frameworks may significantly improve existing communication and control techniques for multi-agent tasks by implicitly representing not only information associated with the individual agents, such as agent position, velocity, and camera data, but also their relationships with one another, such as distances between the agents and their ability to communicate with one another. One such task is a multi-agent navigation problem in which the agents must coordinate with one another in a decentralized manner, using proximity sensors only, to navigate safely to their intended goal positions in the environment without collisions or deadlocks. The contribution of this thesis is the design of an end-to-end decentralized control scheme for multi-agent navigation that utilizes GNNs to prevent inter-agent collisions and deadlocks. The contributions consist of the development, simulation and evaluation of the performance of an advantage actor-critic (A2C) reinforcement learning algorithm that employs actor and critic networks for training that simultaneously approximate the policy function and value function, respectively. These networks are implemented using GNN frameworks for navigation by groups of 3, 5, 10 and 15 agents in simulated two-dimensional environments. It is observed that in $40\%$ to $50\%$ of the simulation trials, between 70$\%$ to 80$\%$ of the agents reach their goal positions without colliding with other agents or becoming trapped in deadlocks. The model is also compared to a random run simulation, where actions are chosen randomly for the agents and observe that the model performs notably well for smaller groups of agents.
ContributorsAyalasomayajula, Manaswini (Author) / Berman, Spring (Thesis advisor) / Mian, Sami (Committee member) / Pavlic, Theodore (Committee member) / Arizona State University (Publisher)
Created2022
Description
Autonomous systems inevitably must interact with other surrounding systems; thus, algorithms for intention/behavior estimation are of great interest. This thesis dissertation focuses on developing passive and active model discrimination algorithms (PMD and AMD) with applications to set-valued intention identification and fault detection for uncertain/bounded-error dynamical systems. PMD uses the obtained input-output data to invalidate the models, while AMD designs an auxiliary input to assist the discrimination process. First, PMD algorithms are proposed for noisy switched nonlinear systems constrained by metric/signal temporal logic specifications, including systems with lossy data modeled by (m,k)-firm constraints. Specifically, optimization-based algorithms are introduced for analyzing the detectability/distinguishability of models and for ruling out models that are inconsistent with observations at run time. On the other hand, two AMD approaches are designed for noisy switched nonlinear models and piecewise affine inclusion models, which involve bilevel optimization with integer variables/constraints in the inner/lower level. The first approach solves the inner problem using mixed-integer parametric optimization, whose solution is included when solving the outer problem/higher level, while the second approach moves the integer variables/constraints to the outer problem in a manner that retains feasibility and recasts the problem as a tractable mixed-integer linear programming (MILP). Furthermore, AMD algorithms are proposed for noisy discrete-time affine time-invariant systems constrained by disjunctive and coupled safety constraints. To overcome the issues associated with generalized semi-infinite constraints due to state-dependent input constraints and disjunctive safety constraints, several constraint reformulations are proposed to recast the AMD problems as tractable MILPs. Finally, partition-based AMD approaches are proposed for noisy discrete-time affine time-invariant models with model-independent parameters and output measurement that are revealed at run time. Specifically, algorithms with fixed and adaptive partitions are proposed, where the latter improves on the performance of the former by allowing the partitions to be optimized. By partitioning the operation region, the problem is solved offline, and partition trees are constructed which can be used as a `look-up table' to determine the optimal input depending on revealed information at run time.
ContributorsNiu, Ruochen (Author) / Yong, Sze Zheng S.Z. (Thesis advisor) / Berman, Spring (Committee member) / Ren, Yi (Committee member) / Zhang, Wenlong (Committee member) / Zhuang, Houlong (Committee member) / Arizona State University (Publisher)
Created2022
Description
Tire blowout often occurs during driving, which can suddenly disturb vehicle motions and seriously threaten road safety. Currently, there is still a lack of effective methods to mitigate tire blowout risks in everyday traffic, even for automated vehicles. To fundamentally study and systematically resolve the tire blowout issue for automated vehicles, a collaborative project between General Motors (GM) and Arizona State University (ASU) has been conducted since 2018. In this dissertation, three main contributions of this project will be presented. First, to explore vehicle dynamics with tire blowout impacts and establish an effective simulation platform for close-loop control performance evaluation, high-fidelity tire blowout models are thoroughly developed by explicitly considering important vehicle parameters and variables. Second, since human cooperation is required to control Level 2/3 partially automated vehicles (PAVs), novel shared steering control schemes are specifically proposed for tire blowout to ensure safe vehicle stabilization via cooperative driving. Third, for Level 4/5 highly automated vehicles (HAVs) without human control, the development of control-oriented vehicle models, controllability study, and automatic control designs are performed based on impulsive differential systems (IDS) theories.
Co-simulations Matlab/Simulink® and CarSim® are conducted to validate performances of all models and control designs proposed in this dissertation. Moreover, a scaled test vehicle at ASU and a full-size test vehicle at GM are well instrumented for data collection and control implementation. Various tire blowout experiments for different scenarios are conducted for more rigorous validations. Consequently, the proposed high-fidelity tire blowout models can correctly and more accurately describe vehicle motions upon tire blowout. The developed shared steering control schemes for PAVs and automatic control designs for HAVs can effectively stabilize a vehicle to maintain path following performance in the driving lane after tire blowout.
In addition to new research findings and developments in this dissertation, a pending patent for tire blowout detection is also generated in the tire blowout project. The obtained research results have attracted interest from automotive manufacturers and could have a significant impact on driving safety enhancement for automated vehicles upon tire blowout.
ContributorsLi, Ao (Author) / Chen, Yan (Thesis advisor) / Berman, Spring (Committee member) / Kannan, Arunachala Mada (Committee member) / Liu, Yongming (Committee member) / Lin, Wen-Chiao (Committee member) / Marvi, Hamidreza (Committee member) / Arizona State University (Publisher)
Created2023
Description
As the explorations beyond the Earth's boundaries continue to evolve, researchers and engineers strive to develop versatile technologies capable of adapting to unknown space conditions. For instance, the utilization of Screw-Propelled Vehicles (SPVs) and robotics that utilize helical screws propulsion to transverse planetary bodies is a growing area of interest. An example of such technology is the Extant Exobiology Life Surveyor (EELS), a snake-like robot currently developed by the NASA Jet Propulsion Laboratory (JPL) to explore the surface of Saturn’s moon, Enceladus. However, the utilization of such a mechanism requires a deep and thorough understanding of screw mobility in uncertain conditions. The main approach to exploring screw dynamics and optimal design involves the utilization of Discrete Element Method (DEM) simulations to assess interactions and behavior of screws when interacting with granular terrains. In this investigation, the Simplified Johnson-Kendall-Roberts (SJKR) model is implemented into the utilized simulation environment to account for cohesion effects similar to what is experienced on celestial bodies like Enceladus. The model is verified and validated through experimental and theoretical testing. Subsequently, the performance characteristics of screws are explored under varying parameters, such as thread depth, number of screw starts, and the material’s cohesion level. The study has examined significant relationships between the parameters under investigation and their influence on the screw performance.
ContributorsAbdelrahim, Mohammad (Author) / Marvi, Hamid (Thesis advisor) / Berman, Spring (Committee member) / Lee, Hyunglae (Committee member) / Arizona State University (Publisher)
Created2023
Description
While wearable soft robots have successfully addressed many inherent design limitations faced by wearable rigid robots, they possess a unique set of challenges due to their soft and compliant nature. Some of these challenges are present in the sensing, modeling, control and evaluation of wearable soft robots. Machine learning algorithms have shown promising results for sensor fusion with wearable robots, however, they require extensive data to train models for different users and experimental conditions. Modeling soft sensors and actuators require characterizing non-linearity and hysteresis, which complicates deriving an analytical model. Experimental characterization can capture the characteristics of non-linearity and hysteresis but requires developing a synthesized model for real-time control. Controllers for wearable soft robots must be robust to compensate for unknown disturbances that arise from the soft robot and its interaction with the user. Since developing dynamic models for soft robots is complex, inaccuracies that arise from the unmodeled dynamics lead to significant disturbances that the controller needs to compensate for. In addition, obtaining a physical model of the human-robot interaction is complex due to unknown human dynamics during walking. Finally, the performance of soft robots for wearable applications requires extensive experimental evaluation to analyze the benefits for the user. To address these challenges, this dissertation focuses on the sensing, modeling, control and evaluation of soft robots for wearable applications. A model-based sensor fusion algorithm is proposed to improve the estimation of human joint kinematics, with a soft flexible robot that requires compact and lightweight sensors. To overcome limitations with rigid sensors, an inflatable soft haptic sensor is developed to enable gait sensing and haptic feedback. Through experimental characterization, a mathematical model is derived to quantify the user's ground reaction forces and the delivered haptic force. Lastly, the performance of a wearable soft exosuit in assisting human users during lifting tasks is evaluated, and the benefits obtained from the soft robot assistance are analyzed.
ContributorsQuiñones Yumbla, Emiliano (Author) / Zhang, Wenlong (Thesis advisor) / Berman, Spring (Committee member) / Lee, Hyunglae (Committee member) / Marvi, Hamid (Committee member) / Sugar, Thomas (Committee member) / Arizona State University (Publisher)
Created2023
Description
This work has improved the quality of the solution to the sparse rewards problemby combining reinforcement learning (RL) with knowledge-rich planning. Classical
methods for coping with sparse rewards during reinforcement learning modify the
reward landscape so as to better guide the learner. In contrast, this work combines
RL with a planner in order to utilize other information about the environment. As
the scope for representing environmental information is limited in RL, this work has
conflated a model-free learning algorithm – temporal difference (TD) learning – with
a Hierarchical Task Network (HTN) planner to accommodate rich environmental
information in the algorithm. In the perpetual sparse rewards problem, rewards
reemerge after being collected within a fixed interval of time, culminating in a lack of a
well-defined goal state as an exit condition to the problem. Incorporating planning in
the learning algorithm not only improves the quality of the solution, but the algorithm
also avoids the ambiguity of incorporating a goal of maximizing profit while using
only a planning algorithm to solve this problem. Upon occasionally using the HTN
planner, this algorithm provides the necessary tweak toward the optimal solution. In
this work, I have demonstrated an on-policy algorithm that has improved the quality
of the solution over vanilla reinforcement learning. The objective of this work has
been to observe the capacity of the synthesized algorithm in finding optimal policies to
maximize rewards, awareness of the environment, and the awareness of the presence
of other agents in the vicinity.
ContributorsNandan, Swastik (Author) / Pavlic, Theodore (Thesis advisor) / Das, Jnaneshwar (Thesis advisor) / Berman, Spring (Committee member) / Arizona State University (Publisher)
Created2022
Description
The Soft Robotic Hip Exosuit (SR-HExo) was designed, fabricated, and tested in treadmill walking experiments with healthy participants to gauge effectivity of the suit in assisting locomotion and in expanding the basin of entrainment as a method of rehabilitation. The SR-HExo consists of modular, compliant materials to move freely with a user’s range of motion and is actuated with X-oriented flat fabric pneumatic artificial muscles (X-ff-PAM) that contract when pressurized and can generate 190N of force at 200kPa in a 0.3 sec window. For use in gait assistance experiments, X-ff-PAM actuators were placed anterior and posterior to the right hip joint. Extension assistance and flexion assistance was provided in 10-45% and 50-90% of the gait cycle, respectively. Device effectivity was determined through range of motion (ROM) preservation and hip flexor and extensor muscular activity reduction. While the active suit reduced average hip ROM by 4o from the target 30o, all monitored muscles experienced significant reductions in electrical activity. The gluteus maximus and biceps femoris experienced electrical activity reduction of 13.1% and 6.6% respectively and the iliacus and rectus femoris experienced 10.7% and 27.7% respectively. To test suit rehabilitative potential, the actuators were programmed to apply periodic torque perturbations to induce locomotor entrainment. An X-ff-PAM was contracted at
the subject’s preferred gait frequency and, in randomly ordered increments of 3%, increased up to 15% beyond.
Perturbations located anterior and posterior to the hip were tested separately to assess impact of location on entrainment characteristics. All 11 healthy participants achieved entrainment in all 12 experimental conditions in both
suit orientations. Phase-locking consistently occurred around toe-off phase of the gait cycle (GC). Extension perturbations synchronized earlier in the gait cycle (before 60% GC where peak hip extension occurs) than flexion perturbations (just after 60% GC at the transition from full hip extension to hip flexion), across group averaged results. The study demonstrated the suit can significantly extend the basin of entrainment and improve transient response compared to previously reported results and confirms that a single stable attractor exists during gait entrainment to unidirectional hip perturbations.
ContributorsBaye-Wallace, Lily (Author) / Lee, Hyunglae (Thesis advisor) / Marvi, Hamidreza (Committee member) / Berman, Spring (Committee member) / Arizona State University (Publisher)
Created2021