Matching Items (8)
Filtering by
- All Subjects: robotics
- All Subjects: Exoskeletons
- Member of: Theses and Dissertations
Description
Coordination and control of Intelligent Agents as a team is considered in this thesis.
Intelligent agents learn from experiences, and in times of uncertainty use the knowl-
edge acquired to make decisions and accomplish their individual or team objectives.
Agent objectives are defined using cost functions designed uniquely for the collective
task being performed. Individual agent costs are coupled in such a way that group ob-
jective is attained while minimizing individual costs. Information Asymmetry refers
to situations where interacting agents have no knowledge or partial knowledge of cost
functions of other agents. By virtue of their intelligence, i.e., by learning from past
experiences agents learn cost functions of other agents, predict their responses and
act adaptively to accomplish the team’s goal.
Algorithms that agents use for learning others’ cost functions are called Learn-
ing Algorithms, and algorithms agents use for computing actuation (control) which
drives them towards their goal and minimize their cost functions are called Control
Algorithms. Typically knowledge acquired using learning algorithms is used in con-
trol algorithms for computing control signals. Learning and control algorithms are
designed in such a way that the multi-agent system as a whole remains stable during
learning and later at an equilibrium. An equilibrium is defined as the event/point
where cost functions of all agents are optimized simultaneously. Cost functions are
designed so that the equilibrium coincides with the goal state multi-agent system as
a whole is trying to reach.
In collective load transport, two or more agents (robots) carry a load from point
A to point B in space. Robots could have different control preferences, for example,
different actuation abilities, however, are still required to coordinate and perform
load transport. Control preferences for each robot are characterized using a scalar
parameter θ i unique to the robot being considered and unknown to other robots.
With the aid of state and control input observations, agents learn control preferences
of other agents, optimize individual costs and drive the multi-agent system to a goal
state.
Two learning and Control algorithms are presented. In the first algorithm(LCA-
1), an existing work, each agent optimizes a cost function similar to 1-step receding
horizon optimal control problem for control. LCA-1 uses recursive least squares as
the learning algorithm and guarantees complete learning in two time steps. LCA-1 is
experimentally verified as part of this thesis.
A novel learning and control algorithm (LCA-2) is proposed and verified in sim-
ulations and on hardware. In LCA-2, each agent solves an infinite horizon linear
quadratic regulator (LQR) problem for computing control. LCA-2 uses a learning al-
gorithm similar to line search methods, and guarantees learning convergence to true
values asymptotically.
Simulations and hardware implementation show that the LCA-2 is stable for a
variety of systems. Load transport is demonstrated using both the algorithms. Ex-
periments running algorithm LCA-2 are able to resist disturbances and balance the
assumed load better compared to LCA-1.
Intelligent agents learn from experiences, and in times of uncertainty use the knowl-
edge acquired to make decisions and accomplish their individual or team objectives.
Agent objectives are defined using cost functions designed uniquely for the collective
task being performed. Individual agent costs are coupled in such a way that group ob-
jective is attained while minimizing individual costs. Information Asymmetry refers
to situations where interacting agents have no knowledge or partial knowledge of cost
functions of other agents. By virtue of their intelligence, i.e., by learning from past
experiences agents learn cost functions of other agents, predict their responses and
act adaptively to accomplish the team’s goal.
Algorithms that agents use for learning others’ cost functions are called Learn-
ing Algorithms, and algorithms agents use for computing actuation (control) which
drives them towards their goal and minimize their cost functions are called Control
Algorithms. Typically knowledge acquired using learning algorithms is used in con-
trol algorithms for computing control signals. Learning and control algorithms are
designed in such a way that the multi-agent system as a whole remains stable during
learning and later at an equilibrium. An equilibrium is defined as the event/point
where cost functions of all agents are optimized simultaneously. Cost functions are
designed so that the equilibrium coincides with the goal state multi-agent system as
a whole is trying to reach.
In collective load transport, two or more agents (robots) carry a load from point
A to point B in space. Robots could have different control preferences, for example,
different actuation abilities, however, are still required to coordinate and perform
load transport. Control preferences for each robot are characterized using a scalar
parameter θ i unique to the robot being considered and unknown to other robots.
With the aid of state and control input observations, agents learn control preferences
of other agents, optimize individual costs and drive the multi-agent system to a goal
state.
Two learning and Control algorithms are presented. In the first algorithm(LCA-
1), an existing work, each agent optimizes a cost function similar to 1-step receding
horizon optimal control problem for control. LCA-1 uses recursive least squares as
the learning algorithm and guarantees complete learning in two time steps. LCA-1 is
experimentally verified as part of this thesis.
A novel learning and control algorithm (LCA-2) is proposed and verified in sim-
ulations and on hardware. In LCA-2, each agent solves an infinite horizon linear
quadratic regulator (LQR) problem for computing control. LCA-2 uses a learning al-
gorithm similar to line search methods, and guarantees learning convergence to true
values asymptotically.
Simulations and hardware implementation show that the LCA-2 is stable for a
variety of systems. Load transport is demonstrated using both the algorithms. Ex-
periments running algorithm LCA-2 are able to resist disturbances and balance the
assumed load better compared to LCA-1.
ContributorsKAMBAM, KARTHIK (Author) / Zhang, Wenlong (Thesis advisor) / Nedich, Angelia (Thesis advisor) / Ren, Yi (Committee member) / Arizona State University (Publisher)
Created2018
Description
Many industries require workers in warehouse and stockroom environments to perform frequent lifting tasks. Over time these repeated tasks can lead to excess strain on the worker's body and reduced productivity. This project seeks to develop an exoskeletal wrist fixture to be used in conjunction with a powered exoskeleton arm to aid workers performing box lifting types of tasks. Existing products aimed at improving worker comfort and productivity typically employ either fully powered exoskeleton suits or utilize minimally powered spring arms and/or fixtures. These designs either reduce stress to the user's body through powered arms and grippers operated via handheld controls which have limited functionality, or they use a more minimal setup that reduces some load, but exposes the user's hands and wrists to injury by directing support to the forearm. The design proposed here seeks to strike a balance between size, weight, and power requirements and also proposes a novel wrist exoskeleton design which minimizes stress on the user's wrists by directly interfacing with the object to be picked up. The design of the wrist exoskeleton was approached through initially selecting degrees of freedom and a ROM (range of motion) to accommodate. Feel and functionality were improved through an iterative prototyping process which yielded two primary designs. A novel "clip-in" method was proposed to allow the user to easily attach and detach from the exoskeleton. Designs utilized a contact surface intended to be used with dry fibrillary adhesives to maximize exoskeleton grip. Two final designs, which used two pivots in opposite kinematic order, were constructed and tested to determine the best kinematic layout. The best design had two prototypes created to be worn with passive test arms that attached to the user though a specially designed belt.
ContributorsGreason, Kenneth Berend (Author) / Sugar, Thomas (Thesis director) / Holgate, Matthew (Committee member) / Mechanical and Aerospace Engineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2016-12
Description
This thesis details the design and construction of a torque-controlled robotic gripper for use with the Pheeno swarm robotics platform. This project required expertise from several fields of study including: robotic design, programming, rapid prototyping, and control theory. An electronic Inertial Measurement Unit and a DC Motor were both used along with 3D printed plastic components and an electronic motor control board to develop a functional open-loop controlled gripper for use in collective transportation experiments. Code was developed that effectively acquired and filtered rate of rotation data alongside other code that allows for straightforward control of the DC motor through experimentally derived relationships between the voltage applied to the DC motor and the torque output of the DC motor. Additionally, several versions of the physical components are described through their development.
ContributorsMohr, Brennan (Author) / Berman, Spring (Thesis director) / Ren, Yi (Committee member) / Mechanical and Aerospace Engineering Program (Contributor) / School for Engineering of Matter,Transport & Enrgy (Contributor) / Barrett, The Honors College (Contributor)
Created2019-05
Description
For the past two decades, advanced Limb Gait Simulators and Exoskeletons have been developed to improve walking rehabilitation. A Limb Gait Simulator is used to analyze the human step cycle and/or assist a user walking on a treadmill. Most modern limb gait simulators, such as ALEX, have proven themselves effective and reliable through their usage of motors, springs, cables, elastics, pneumatics and reaction loads. These mechanisms apply internal forces and reaction loads to the body. On the other hand, external forces are those caused by an external agent outside the system such as air, water, or magnets. A design for an exoskeleton using external forces has seldom been attempted by researchers. This thesis project focuses on the development of a Limb Gait Simulator based on a Pure External Force and has proven its effectiveness in generating torque on the human leg. The external force is generated through air propulsion using an Electric Ducted Fan (EDF) motor. Such a motor is typically used for remote control airplanes, but their applications can go beyond this. The objective of this research is to generate torque on the human leg through the control of the EDF engines thrust and the opening/closing of the reverse thruster flaps. This device qualifies as "assist as needed"; the user is entirely in control of how much assistance he or she may want. Static thrust values for the EDF engine are recorded using a thrust test stand. The product of the thrust (N) and the distance on the thigh (m) is the resulting torque. With the motor running at maximum RPM, the highest torque value reached was that of 3.93 (Nm). The motor EDF motor is powered by a 6S 5000 mAh LiPo battery. This torque value could be increased with the usage of a second battery connected in series, but this comes at a price. The designed limb gait simulator demonstrates that external forces, such as air, could have potential in the development of future rehabilitation devices.
ContributorsToulouse, Tanguy Nathan (Author) / Sugar, Thomas (Thesis director) / Artemiadis, Panagiotis (Committee member) / Mechanical and Aerospace Engineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2016-12
Description
In order for assistive mobile robots to operate in the same environment as humans, they must be able to navigate the same obstacles as humans do. Many elements are required to do this: a powerful controller which can understand the obstacle, and power-dense actuators which will be able to achieve the necessary limb accelerations and output energies. Rapid growth in information technology has made complex controllers, and the devices which run them considerably light and cheap. The energy density of batteries, motors, and engines has not grown nearly as fast. This is problematic because biological systems are more agile, and more efficient than robotic systems. This dissertation introduces design methods which may be used optimize a multiactuator robotic limb's natural dynamics in an effort to reduce energy waste. These energy savings decrease the robot's cost of transport, and the weight of the required fuel storage system. To achieve this, an optimal design method, which allows the specialization of robot geometry, is introduced. In addition to optimal geometry design, a gearing optimization is presented which selects a gear ratio which minimizes the electrical power at the motor while considering the constraints of the motor. Furthermore, an efficient algorithm for the optimization of parallel stiffness elements in the robot is introduced. In addition to the optimal design tools introduced, the KiTy SP robotic limb structure is also presented. Which is a novel hybrid parallel-serial actuation method. This novel leg structure has many desirable attributes such as: three dimensional end-effector positioning, low mobile mass, compact form-factor, and a large workspace. We also show that the KiTy SP structure outperforms the classical, biologically-inspired serial limb structure.
ContributorsCahill, Nathan M (Author) / Sugar, Thomas (Thesis advisor) / Ren, Yi (Thesis advisor) / Holgate, Matthew (Committee member) / Berman, Spring (Committee member) / Artemiadis, Panagiotis (Committee member) / Arizona State University (Publisher)
Created2017
Description
Bicycle stabilization has become a popular topic because of its complex dynamic behavior and the large body of bicycle modeling research. Riding a bicycle requires accurately performing several tasks, such as balancing and navigation which may be difficult for disabled people. Their problems could be partially reduced by providing steering assistance. For stabilization of these highly maneuverable and efficient machines, many control techniques have been applied – achieving interesting results, but with some limitations which includes strict environmental requirements. This thesis expands on the work of Randlov and Alstrom, using reinforcement learning for bicycle self-stabilization with robotic steering. This thesis applies the deep deterministic policy gradient algorithm, which can handle continuous action spaces which is not possible for Q-learning technique. The research involved algorithm training on virtual environments followed by simulations to assess its results. Furthermore, hardware testing was also conducted on Arizona State University’s RISE lab Smart bicycle platform for testing its self-balancing performance. Detailed analysis of the bicycle trial runs are presented. Validation of testing was done by plotting the real-time states and actions collected during the outdoor testing which included the roll angle of bicycle. Further improvements in regard to model training and hardware testing are also presented.
ContributorsTurakhia, Shubham (Author) / Zhang, Wenlong (Thesis advisor) / Yong, Sze Zheng (Committee member) / Ren, Yi (Committee member) / Arizona State University (Publisher)
Created2020
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
Recent years, there has been many attempts with different approaches to the human-robot interaction (HRI) problems. In this paper, the multi-agent interaction is formulated as a differential game with incomplete information. To tackle this problem, the parameter estimation method is utilized to obtain the approximated solution in a real time basis. Previous studies in the parameter estimation made the assumption that the human parameters are known by the robot; but such may not be the case and there exists uncertainty in the modeling of the human rewards as well as human's modeling of the robot's rewards. The proposed method, empathetic estimation, is tested and compared with the ``non-empathetic'' estimation from the existing works. The case studies are conducted in an uncontrolled intersection with two agents attempting to pass efficiently. Results have shown that in the case of both agents having inconsistent belief of the other agent's parameters, the empathetic agent performs better at estimating the parameters and has higher reward values, which indicates the scenarios when empathy is essential: when agent's initial belief is mismatched from the true parameters/intent of the agents.
ContributorsChen, Yi (Author) / Ren, Yi (Thesis advisor) / Zhang, Wenlong (Committee member) / Yong, Sze Zheng (Committee member) / Arizona State University (Publisher)
Created2021