DOC PREVIEW
FUZZY LOGIC CONTROL

This preview shows page 1-2 out of 7 pages.

Save
View full document
View full document
Premium Document
Do you want full access? Go Premium and unlock all 7 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 7 pages.
Access to all documents
Download any document
Ad free experience
Premium Document
Do you want full access? Go Premium and unlock all 7 pages.
Access to all documents
Download any document
Ad free experience

Unformatted text preview:

1 Copyright ©2004 ASME Proceedings of DETC’04: ASME 2004 Design Engineering Technical Conferences and Computers and Information in Engineering Conference September 28-October 2, 2004, Salt Lake City, Utah USA DETC2004 – 57723 FUZZY LOGIC CONTROL FOR AN AUTONOMOUS UNDERWATER CRAWLER Douglas Welling / University of Idaho Dean Edwards / University of Idaho Mike Anderson / University of Idaho ABSTRACT Current underwater crawling vehicles could benefit by using rotating head sonar data to avoid collisions with obstacles. We have developed and optimized a fuzzy logic controller using software for simulation of an underwater environment. The optimization results show near an order of magnitude increase in performance over both straight line and lawnmower search patterns with relatively small changes in the system parameters. The fuzzy logic controller has the capability of navigating a crawler safely and quickly between mission specific points. Keywords: Fuzzy Logic, Autonomous, Optimal Control. INTRODUCTION Even in a time of relative world peace there is still a need to maintain the largest and most effective navy in the world. Autonomous mine countermeasures is a relatively old idea that is just recently being realized. Single autonomous vehicles or fleets of autonomous vehicles are being used to search for and classify mines in shallow water and surf zones. Some applications for this process are shallow water warfare, amphibious assaults, mine reconnaissance, and terrain identification. Such vehicles have been used successfully in Operation Iraqi Freedom, and are being commissioned by the Navy to serve in future operations. Crawling Autonomous Underwater Vehicles (AUVs) are referred to as crawlers (Figure 1). Underwater crawlers are currently under development for use by the Navy. The purpose of a crawler is to sweep areas of very shallow water, surf zones, and beaches while detecting and classifying mines and non-threats. Lawnmower style search patterns are typically performed. A crawler will receive position updates from Long BaseLine (LBL) acoustic transponders while in the water and can receive Global Positioning System (GPS) positions when on shore. LBL transponders, at least two, are placed up to a mile away from the crawler’s location and can be used to triangulate vehicle position. Current crawler control systems use modern control theory to move between mission specific locations (referred to as waypoints). In the event that there is an obstacle between two waypoints, current vehicles use a “bumper” to determine when contact with an obstacle has been made. The vehicle then does a predetermined set of movements to avoid the object and continue back on course. This control method works well for sparsely populated obstacle fields. In the presence of more obstacles or Counter Counter Measures (CCMs), such as net lines, pits, or wire traps used to disable a crawler, the use of a “bumper” for obstacle avoidance is not optimal. Densely populated obstacle fields make a pre-determined movement pattern unreliable since it is highly probable that the vehicle will encounter many obstacles Figure 1 – Example of an Underwater Crawler2 Copyright ©2004 ASME in the attempt to avoid one. Letting a vehicle drive into a CCM will likely trap the vehicle and effectively end its mission. The University of Idaho has developed a way to control crawlers through obstacle fields. A fuzzy logic control has been developed and optimized that uses data from a rotating head sonar to safely and quickly navigate between waypoints. The controller has been developed in a simulated environment and optimized using the simplex method. VEHICLE Crawlers are typically tracked vehicles measuring about 24” in width, 36” in length, with the highest point extending up to 36”. The vehicle is propelled by an electric motor using battery packs for energy storage. The sensor package can vary for each specific vehicle and mission [1]. The simulated vehicle used shaft encoders, a magnetic compass, and rotating head sonar. No mine detection or classification was done in this simulation. Shaft encoders provide motion feedback from the motors and when used along with the magnetic compass provides the vehicle’s location. Such internal navigation systems can accumulate large bias errors. For the simulation, it was assumed that the vehicle receives frequent updates from an LBL system so that bias errors can be neglected. Rotating head sonar can operate in frequencies between 300 kHz and 1 MHz with a maximum range of 200 meters. The simulation assumed a range of 20 meters and that the returned data consisted of the distance to the nearest object for each degree of head rotation. A typical rotating head sonar is shown in Figure 2. ENVIRONMENT AND SIMULATION The software used for the simulation environment is the Autonomous Littoral Warfare Systems Evaluator – Monte Carlo (ALWSE-MC) developed and maintained by Naval Surface Warfare Center Panama City [2]. ALWSE-MC is a kinematic, statistical AUV mission simulator. Mine fields are created with obstacles and CCMs. Vehicles are simulated by a point object using customizable sensor packages and kinematic constraints. Waypoints can be defined, and underwater acoustic communication is simulated to give position updates. Many other simulation options are available inside ALWSE-MC. ALWSE-MC has a behavior module option that when activated will execute Matlab control script every simulation time step. Our algorithms were written in Matlab script and run in ALWSE-MC in this manner. CONTROLS A hierarchical fuzzy logic control system has been chosen that uses behavior modules. The fuzzy logic control system was designed to exhibit both obstacle avoidance and path finding behaviors see Figure 3. When a vehicle approaches an obstacle, it uses the rotating head sonar data to determine the most efficient route around the obstacle and then finds the best way back to the original path. The idea is to have the vehicle travel in a straight line between waypoints with minimal deviation. The vehicle only deviates from the path for short distances to move around an obstacle and return to the path. The ability of the vehicle to stay on a straight line path helps ensure consistent area coverage over multiple passes either by one or multiple vehicles. Fuzzy logic is a superset of conventional logic that has


FUZZY LOGIC CONTROL

Download FUZZY LOGIC CONTROL
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view FUZZY LOGIC CONTROL and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view FUZZY LOGIC CONTROL 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?