This thesis addresses two important aspects of automatic assembly viz., assembly sequence planning and assembly path planning. These issues are addressed separately starting with sequence planning followed by assembly path planning. For efficient assembly without feedback systems (or, passive assembly), an assembler should know the ideal orientation of each component and the order in which to put the parts together (or, assembly sequence). A heuristic is presented to find the optimal assembly sequence and prescribe the orientation of the components for a minimum set of grippers = ideally one. The heuristic utilizes an index of difficulty (ID) that quantifies assembly. The ID for each task in the assembly process is computed on the basis of a number of geometrical and operational properties. The objective of the optimization problem here is to minimize the assembly ID and categorize parts/subassemblies based on their preferred direction of assembly while allowing re-orientation of the base part. It is assumed that the preferred direction of assembly is vertically downward, consistent with manual as well as most automatic assembly protocols. Our attempt is to minimize the number of degrees of freedom required in a re-orienting fixture and derive the requirements for such a fixture. The assembly of a small engine is used as an example in this study due to the variety of ideally rigid parts involved. In high precision assembly tasks, contact motion is common and often desirable. This entails a careful study of contact states of the parts being assembled. Recognition of contact states is crucial in planning and executing contact motion plans due to inevitable uncertainties. Dr. Jing Xiao of UNCC introduced the concept of principal contacts (PC) and contact formation (CF) for contact state recognition. The concept of using CFs (as sets of PCs) has the inherent advantage that a change of CF is often coincident with a discontinuity of the general contact force (force and torque). Previous work in contact motion planning has shown that contact information at the level of PCs along with the sensed location and force information is often sufficient for planning high precision assembly operations. In this thesis, we present results from experiments involving planned contact motions to validate the notion of PCs and CFs -- an abrupt change in general contact force often accompanies a change between CFs. We are only concerned with solving the 2D peg-in-corner problem. / Master of Science

Most commercial CAD systems do not offer sufficient support for the design activity. The reason is that they cannot understand the functional requirements of the design product. The user is responsible for maintaining the functional requirements in different design phases. By incorporating constraint programming concepts, these CAD systems would evolve into systems which would maintain the functional requirements in the design process, and perform analysis and simulation of geometric models. The CAD systems incorporated with constraint programming concepts would reduce design time, avoid human fatigue and error, and also maintain consistency of the geometric constraints imposed on the model. The G+ system addresses these issues by introducing a constraint-based system for geometric modeling by object-oriented methods. The G+ is designed such that available specialized algorithms can be utilized to enable handling of non-linear problems by both iterative and non-iterative schemes.

No description available.

No description available.

No description available.

No description available.

No description available.

This dissertation is an investigation and analysis of the contributions of the human organism to climate change, specifically concerned with its adverse effect on the biosphere. The focus of the analysis rests on three phenomena distinctive to humans participating in globalized economies and modernized societies; the distortions of anthrosupremacy, an uncritical development of technology, and an unhealthy cathexis with economic growth. These are analyzed and identified as stark contributors to the creation, proliferation, and pervasiveness of systems contributing to the decline of the biosphere. These phenomena serve to dissociate humans from the consequences of their actions in systemic ways; thereby thriving by preying on the faults of human agency. Using a biogenic framework to analyze human agency, these faults are identified as problematic to the teleological pursuits of the human organism (i.e., extantcy). An analysis of these phenomena in relationship to the faults in human agency is done to create an awareness of how certain mythologies, technologies, and socioeconomic practices have created enabling constraints that deter a more efficacious and sustainable exercise of human agency. Existing strategies and philosophies that have been promoted to address the issue have failed to account for these enabling constraints and, have therefore, fallen short of creating their desired effect in practical ways. This dissertation concludes that, without altering these enabling constraints to allow for a more efficacious and sustainable exercise of agency, the human species is less likely to persist into the indefinite future.

In recent years, the number of exploits targeting software applications has increased dramatically. These exploits have caused substantial economic damages. Ensuring that software applications are not vulnerable to the exploits has, therefore, become a critical requirement. The last line of defense is to test before hand if a software application is vulnerable to exploits. One can accomplish this by testing for the presence of vulnerabilities. This dissertation presents a framework for deriving verification and validation (V&V) strategies to assess the security of a software application by testing it for the presence of vulnerabilities. This framework can be used to assess the security of any software application that executes above the level of the operating system. It affords a novel approach, which consists of testing if the software application permits violation of constraints imposed by computer system resources or assumptions made about the usage of these resources. A vulnerability exists if a constraint or an assumption can be violated. Distinctively different from other approaches found in the literature, this approach simplifies the process of assessing the security of a software application. The framework is composed of three components: (1) a taxonomy of vulnerabilities, which is an informative classification of vulnerabilities, where vulnerabilities are expressed in the form of violable constraints and assumptions; (2) an object model, which is a collection of potentially vulnerable process objects that can be present in a software application; and (3) a V&V strategies component, which combines information from the taxonomy and the object model; and provides approaches for testing software applications for the presence of vulnerabilities. This dissertation also presents a step-by-step process for using the framework to assess software security. / Ph. D.

There is an increasing need for low-power neural systems as neural networks become more widely used in embedded devices with limited resources. Spiking neural networks (SNNs) are proving to be a more energy-efficient option to conventional Artificial neural networks (ANNs), which are recognized for being computationally heavy. Despite its significance, there has been not enough attention on training SNNs on large-scale distributed Machine Learning techniques like Federated Learning (FL). As federated learning involves many energy-constrained devices, there is a significant opportunity to take advantage of the energy efficiency offered by SNNs. However, it is necessary to address the real-world communication constraints in an FL system and this is addressed with the help of three communication reduction techniques, namely, model compression, partial device participation, and periodic aggregation. Furthermore, the convergence of federated learning systems is also affected by data heterogeneity. Federated learning systems are capable of protecting the private data of clients from adversaries. However, by analyzing the uploaded client parameters, confidential information can still be revealed. To combat privacy attacks on the FL systems, various attempts have been made to incorporate differential privacy within the framework. In this thesis, we investigate the trade-offs between communication costs and training variance under a Federated Learning system with Differential Privacy applied at the parameter server (curator model). / Master of Science / Federated Learning is a decentralized method of training neural network models; it employs several participating devices to independently learn a model on their local data partition. These local models are then aggregated at a central server to achieve the same performance as if the model had been trained centrally. But with Federated Learning systems there is a communication overhead accumulated. Various communication reductions can be used to reduce these costs. Spiking Neural Networks, being the energy-efficient option to Artificial Neural Networks, can be utilized in Federated Learning systems. This is because FL systems consist of a network of energy-efficient devices. Federated learning systems are helpful in preserving the privacy of data in the system. However, an attacker can still obtain meaningful information from the parameters that are transmitted during a session. To this end, differential privacy techniques are utilized to combat privacy concerns in Federated Learning systems. In this thesis, we compare and contrast different communication costs and parameters of a federated learning system with differential privacy applied to it.