Matching Items (5)
Filtering by
- All Subjects: Optimization
- Creators: Zhang, Wenlong
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
The global population over the age of 60 is estimated to rise to 23% by 2050 only increase the prevalence of functional neurological disorders and stroke. Increase in cases of functional neurological disorders and strokes will place a greater burden on the healthcare industry, specifically physical therapy. Physical therapy is vital for a patient’s recovery of motor function which is time demanding and taxing on the physical therapist. Wearable robotics have been proven to improve functional outcomes in gait rehabilitation by providing controlled high dosage and high-intensity training. Accurate control strategies for assistive robotic exoskeletons are vital for repetitive high precisions assistance for cerebral plasticity to occur.
This thesis presents a preliminary determination and design of a control algorithm for an assistive ankle device developed by the ASU RISE Laboratory. The assistive ankle device functions by compressing a spring upon heel strike during gait, remaining compressed during mid-stance and then releasing upon initiation of heel-off. The relationship between surface electromyography and ground reactions forces were used for identification of user-initiated heel-off. The muscle activation of the tibialis anterior combined with the ground reaction forces of the heel pressure sensor generated potential features that will be utilized in the revised control algorithm for the assistive ankle device. Work on this project must proceed in order to test and validate the revised control algorithm to determine its accuracy and precision.
This thesis presents a preliminary determination and design of a control algorithm for an assistive ankle device developed by the ASU RISE Laboratory. The assistive ankle device functions by compressing a spring upon heel strike during gait, remaining compressed during mid-stance and then releasing upon initiation of heel-off. The relationship between surface electromyography and ground reactions forces were used for identification of user-initiated heel-off. The muscle activation of the tibialis anterior combined with the ground reaction forces of the heel pressure sensor generated potential features that will be utilized in the revised control algorithm for the assistive ankle device. Work on this project must proceed in order to test and validate the revised control algorithm to determine its accuracy and precision.
ContributorsGaytan-Jenkins, Daniel Rinaldo (Author) / Zhang, Wenlong (Thesis director) / Tyler, Jamie (Committee member) / Harrington Bioengineering Program (Contributor, Contributor) / Barrett, The Honors College (Contributor)
Created2019-05
Description
The thesis covers the development and modeling of the supervisory hybrid controller using two different methods to achieve real-world optimization and power split of a parallel hybrid vehicle with a fixed shaft connecting the Internal Combustion Engine (ICE) and Electric Motor (EM). The first strategy uses a rule based controller to determine modes the vehicle should operate in. This approach is well suited for real-world applications. The second approach uses Sequential Quadratic Programming (SQP) approach in conjunction with an Equivalent Consumption Minimization Strategy (ECMS) strategy to keep the vehicle in the most efficient operating regions. This latter method is able to operate the vehicle in various drive cycles while maintaining the SOC with-in allowed charge sustaining (CS) limits. Further, the overall efficiency of the vehicle for all drive cycles is increased. The limitation here is the that process is computationally expensive; however, with advent of the low cost high performance hardware this method can be used for the hybrid vehicle control.
ContributorsMaady, Rashad Kamal (Author) / Redkar, Sangram (Thesis advisor) / Mayyas, Abdel R (Thesis advisor) / Zhang, Wenlong (Committee member) / Arizona State University (Publisher)
Created2017
Description
Multibody Dynamic (MBD) models are important tools in motion analysis and are used to represent and accurately predict the behavior of systems in the real-world. These models have a range of applications, including the stowage and deployment of flexible deployables on spacecraft, the dynamic response of vehicles in automotive design and crash testing, and mapping interactions of the human body. An accurate model can aid in the design of a system to ensure the system is effective and meets specified performance criteria when built. A model may have many design parameters, such as geometrical constraints and component mechanical properties, or controller parameters if the system uses an external controller. Varying these parameters and rerunning analyses by hand to find an ideal design can be time consuming for models that take hours or days to run. To reduce the amount of time required to find a set of parameters that produces a desired performance, optimization is necessary. Many papers have discussed methods for optimizing rigid and flexible MBD models, and separately their controllers, using both gradient-based and gradient-free algorithms. However, these optimization methods have not been used to optimize full-scale MBD models and their controllers simultaneously. This thesis presents a method for co-optimizing an MBD model and controller that allows for the flexibility to find model and controller-based solutions for systems with tightly coupled parameters. Specifically, the optimization is performed on a quadrotor drone MBD model undergoing disturbance from a slung load and its position controller to meet specified position error performance criteria. A gradient-free optimization algorithm and multiple objective approach is used due to the many local optima from the tradeoffs between the model and controller parameters. The thesis uses nine different quadrotor cases with three different position error formulations. The results are used to determine the effectiveness of the optimization and the ability to converge on a single optimal design. After reviewing the results, the optimization limitations are discussed as well as the ability to transition the optimization to work with different MBD models and their controllers.
ContributorsGambatese, Marcus (Author) / Zhang, Wenlong (Thesis advisor) / Berman, Spring (Committee member) / Inoyama, Daisaku (Committee member) / Arizona State University (Publisher)
Created2022
Description
This dissertation develops advanced controls for distributed energy systems and evaluates performance on technical and economic benefits. Microgrids and thermal systems are of primary focus with applications shown for residential, commercial, and military applications that have differing equipment, rate structures, and objectives. Controls development for residential energy heating and cooling systems implement adaptive precooling strategies and thermal energy storage, with comparisons made of each approach separately and then together with precooling and thermal energy storage. Case studies show on-peak demand and annual energy related expenses can be reduced by up to 75.6% and 23.5%, respectively, for a Building America B10 Benchmark home in Phoenix Arizona, Los Angeles California, and Kona Hawaii. Microgrids for commercial applications follow after with increased complexity. Three control methods are developed and compared including a baseline logic-based control, model predictive control, and model predictive control with ancillary service control algorithms. Case studies show that a microgrid consisting of 326 kW solar PV, 634 kW/ 634 kWh battery, and a 350 kW diesel generator can reduce on-peak demand and annual energy related expenses by 82.2% and 44.1%, respectively. Findings also show that employing a model predictive control algorithm with ancillary services can reduce operating expenses by 23.5% when compared to a logic-based algorithm. Microgrid evaluation continues with an investigation of off-grid operation and resilience for military applications. A statistical model is developed to evaluate the survivability (i.e. probability to meet critical load during an islanding event) to serve critical load out to 7 days of grid outage. Case studies compare the resilience of a generator-only microgrid consisting of 5,250 kW in generators and hybrid microgrid consisting of 2,250 kW generators, 3,450 kW / 13,800 kWh storage, and 16,479 kW solar photovoltaics. Findings show that the hybrid microgrid improves survivability by 10.0% and decreases fuel consumption by 47.8% over a 168-hour islanding event when compared to a generator-only microgrid under nominal conditions. Findings in this dissertation can increase the adoption of reliable, low cost, and low carbon distributed energy systems by improving the operational capabilities and economic benefits to a variety of customers and utilities.
ContributorsNelson, James Robert (Author) / Johnson, Nathan (Thesis advisor) / Stadler, Michael (Committee member) / Zhang, Wenlong (Committee member) / Arizona State University (Publisher)
Created2019