Nicholas Ernest

Fuzzy Clustering based Genetic Algorithm for the Multi-Depot Polygon Visiting Dubins Multiple Traveling Salesman Problem

A genetic algorithm (GA) and Fuzzy Logic System (FLS) based approach to the path representation of a variant of the Traveling Salesman Problem (TSP), the Multi-Depot Polygon Visiting Dubins Multiple Traveling Salesman Problem (MDPVDMTSP) is presented. Utilizing a hybridization of control techniques, this work effectively and efficiently approximates path planning and visbility problems encountered by a UAV swarm in the constant altitude, constant velocity, two-dimensional case. Benchmarking capabilities only exist for a 20 polygon PVDTSP case (1 UAV), but for this case a 9.8% increase in accuracy along with an order of magnitude decrease in run-time was found compared to the alternative method, despite utilizing a slower computer and programming language. Over 100 runs, the best solution was found 99% of the times, with a 1% chance of settling to a local minima. Comparison opportunities for MDPVDMTSP’s are currently not present, but the algorithms work similarly well for this scenario. MDPVDMTSP’s with 250 polygons, 20 UAV’s, and 4 different depots can be accurately approximated in under 30 seconds on the same machine. Being a combination of approximate methods, a decrease in run-time was expected, however the accuracy of this work shows great future promise for more complex variants of this problem.

Self-Crossover Based Genetic Algorithm for Performance Augmentation of the Traveling Salesman Problem

A genetic algorithm (GA) based approach to the path representation of the Traveling Salesman Problem (TSP) will be presented towards potential application to co-operative control of a group of UAVs. The scenario involves a random distribution of cities on a two dimensional grid or a pre-defined set of targets, and determines the optimal path solution of the TSP by means of an iterative algorithm. Similar codes based on genetic algorithms are in existence; however, a unique set of modifications are utilized in this study. Generally speaking, when applying a GA to the TSP, traditional crossover is not permitted, as a city could be visited twice using this method. To circumvent this issue and optimize algorithm efficiency, we introduce a concept referred to as SCROOGE (Self CROssover Optimal GEnetic algorithm). This algorithm will be utilizing a combination of self-crossover and mutation of the strings making up the population. Through selfcrossover, a single string is chosen for breeding, and has a chance to produce an offspring based off of its own genetic material, much as a starfish or many other animals would in nature. Additionally, SCROOGE’s parameters morph with time, becoming more focused on avoiding local minima as iterations progress.

Genetic Fuzzy based Artificial Intelligence for Unmanned Combat AerialVehicle Control in Simulated Air Combat Missions

Breakthroughs in genetic fuzzy systems, most notably the development of the Genetic Fuzzy Tree methodology, have allowed fuzzy logic based Artificial Intelligences to be developed that can be applied to incredibly complex problems. The ability to have extreme performance and computational efficiency as well as to be robust to uncertainties and randomness, adaptable to changing scenarios, verified and validated to follow safety specifications and operating doctrines via formal methods, and easily designed and implemented are just some of the strengths that this type of control brings. Within this white paper, the authors introduce ALPHA, an Artificial Intelligence that controls flights of Unmanned Combat Aerial Vehicles in aerial combat missions within an extreme-fidelity simulation environment. To this day, this represents the most complex application of a fuzzy-logic based Artificial Intelligence to an Unmanned Combat Aerial Vehicle control problem. While development is on-going, the version of ALPHA presented withinwas assessed by Colonel (retired)Gene Lee who described ALPHA as “the most aggressive, responsive, dynamic and credible AI (he’s) seen-to-date.” The quality of these preliminary results in a problem that is not only complex and rife with uncertainties but also contains an intelligent and unrestricted hostile force has significant implications for this type of Artificial Intelligence. This work adds immensely to the body of evidence that this methodology is an ideal solution to a very wide array of problems.

A Hierarchical Market Solution to the Min-Max Multiple Depots Vehicle Routing Problem

The problem of assigning a group of Unmanned Aerial Vehicles (UAVs) to perform spatially distributed tasks often requires that the tasks will be performed as quickly as possible. This problem can be defined as the Min–Max Multiple Depots Vehicle Routing Problem (MMMDVRP), which is a benchmark combinatorial optimization problem. In this problem, UAVs are assigned to service tasks so that each task is serviced once and the goal is to minimize the longest tour performed by any UAV in its motion from its initial location (depot) to the tasks and back to the depot. This problem arises in many time-critical applications, e.g. mobile targets assigned to UAVs in a military context, wildfire fighting, and disaster relief efforts in civilian applications. In this work, we formulate the problem using Mixed Integer Linear Programming (MILP) and Binary Programming and show the scalability limitation of these formulations. To improve scalability, we propose a hierarchical market-based solution (MBS). Simulation results demonstrate the ability of the MBS to solve large scale problems and obtain better costs compared with other known heuristic solution.

Prediction of lithium response in first-episode mania using the LITHium Intelligent Agent (LITHIA): Pilot data and proof-of-concept

Individualized treatment for bipolar disorder based on neuroimaging treatment targets remains elusive. To address this shortcoming, we developed a linguistic machine learning system based on a cascading genetic fuzzy tree (GFT) design called the LITHium Intelligent Agent (LITHIA). Using multiple objectively defined functional magnetic resonance imaging (fMRI) and proton magnetic resonance spectroscopy (1H‐MRS) inputs, we tested whether LITHIA could accurately predict the lithium response in participants with first‐episode bipolar mania.

Genetic Fuzzy Trees and their Application Towards Autonomous Training and Control of a Squadron of Unmanned Combat Aerial Vehicles

This study introduces the technique of Genetic Fuzzy Trees (GFTs) through novel application to an air combat control problem of an autonomous squadron of Unmanned Combat Aerial Vehicles (UCAVs) equ...

Learning of Intelligent Controllers for Autonomous Unmanned Combat Aerial Vehicles by Genetic Cascading Fuzzy Methods

Abstract Looking forward to an autonomous Unmanned Combat Aerial Vehicle (UCAV) for future applications, it becomes apparent that on-board intelligent controllers will be necessary for these advanced systems. LETHA (Learning Enhanced Tactical Handling Algorithm) was created to develop intelligent managers for these advanced unmanned craft through the novel means of a genetic cascading fuzzy system. In this approach, a genetic algorithm creates rule bases and optimizes membership functions for multiple fuzzy logic systems, whose inputs and outputs feed into one another alongside crisp data.A simulation space referred to as HADES (Hoplological Autonomous Defend and Engage Simulation) was created in which LETHA can train the UCAVs intelligent controllers. Equipped with advanced sensors, a limited supply of Self-Defense Missiles (SDM), and a recharging Laser Weapon System (LWS), these UCAVs can navigate a pre-defined route through the mission space, counter enemy threats, and destroy mission-critical targets. Multiple missions were developed in HADES and a squadron of four UCAVs was trained by LETHA. Monte Carlo simulations of the resulting controllers were tested in mission scenarios that are distinct from the training scenarios to determine the training effectiveness in new environments and the presence of deep learning.Despite an incredibly large sample space, LETHA has demonstrated remarkable effectiveness in training intelligent controllers for the UCAV squadron and shown robustness to drastically changing states, uncertainty, and limited information while maintaining extreme levels of computational efficiency. Her specific architecture is applicable to a wide array of topics and specializes in problems with limited distributed resources in a spatiotemporal environment containing uncertainties and unknowns.

Collaborative Tasking of UAV's Using a Genetic Fuzzy Approach

As a continuation of the previous work, under the title “Fuzzy Clustering based Genetic Algorithm for the Multi-Depot Polygon Visiting Dubins Multiple Traveling Salesman Problem”, the techniques described prior [1,2] have since been applied to more complex variants of the Traveling Salesman Problem (TSP). In particular, this paper presents the results of exploring the min-max, or time optimal, variant of the MDPVDMTSP (henceforth referred to as the Min-Max Variant, or MMV), and the Multi-Objective Variant (MOV). Development and integration of genetic fuzzy systems has brought great progress to this research, enabling the algorithms to more closely simulate and solve problems that UAV swarms could encounter. The powerful heuristic methods utilized via multiple genetic algorithms (GA) an fuzzy logic systems (FLS) allow the code to have a very slow climb in computational cost with problem size, while still maintaining a high degree of accuracy.

An Efficient Genetic Fuzzy Approach to UAV Swarm Routing

Fuzzy logic is used in a variety of applications because of its universal approximator attribute and nonlinear characteristics. But, it takes a lot of trial and error to come up with a set of membership functions and rule-base that will effectively work for a specific application. This process could be simplified by using a heuristic search algorithm like Genetic Algorithm (GA). In this paper, genetic fuzzy is applied to the task assignment for cooperating Unmanned Aerial Vehicles (UAVs) classified as the Polygon Visiting Multiple Traveling Salesman Problem (PVMTSP). The PVMTSP finds a lot of applications including UAV swarm routing. We propose a method of genetic fuzzy clustering that would be specific to PVMTSP problems and hence more efficient compared to k-means and c-means clustering. We developed two different algorithms using genetic fuzzy. One evaluates the distance covered by each UAV to cluster the search-space and the other uses a cost function that approximates the distance covered thus resulting in a reduced computational time. We compare these two approaches to each other as well as to an already benchmarked fuzzy clustering algorithm which is the current state-of-the-art. We also discuss how well our algorithm scales for increasing number of targets. The results are compared for small and large polygon sizes.

Multi-agent Cooperative Decision Making using Genetic Cascading Fuzzy Systems

Missions consisting of groups of unmanned aerial vehicles (UAVs) require a high degree of coordination for successfully achieving desired goals. From the many types of applications where a group of coordinated agents is potentially able to outperform a single or a number of systems operating independently, the assignment of tasks is an important one. Different authors have addressed multi-agent task assignment problems for UAVs. For instance, [4] presented a robust task assignment algorithm for uncertain environments. The authors of [1] provided a decentralized task consensus algorithm with asynchronous communication. Decisions to communicate are taken by each agent independently based on the outcomes of assignments using different sets of information. In the present paper we envision a cooperative assignment of tasks in which vehicles decide which threats to attack according to collective preferences, in contrast to individual preferences which has been a common approach in the literature [3, 7–10]. The main objective behind this approach is to encourage agents to make decisions that bring greater benefit to the group. Although this objective may not necessarily suit all assignment problems, in some scenarios like the one presented in this paper the collective approach can potentially lead to an improved cooperation among agents. Additionally, the assignments need to be free of conflicts in order to use resources wisely. In the present work, agents are equipped with different types of weapons which need to be used in optimal manner and according to the specific type of threats. All these different factors need to be considered in order to make good and fast decisions. The main tool used in this paper for each local agent to generate these decisions is based on a novel method of Genetic Cascading Fuzzy Systems [6].