No abstract

A typical Autonomous Ground Robotic Vehicle (AGRV) uses a combination of sensors to monitor movements and the surrounding environment. Placing multiple sensors on an AGRV may allow for complexity in sensor data, but far more important is integration of the information from these multiple sensors to perform a given task optimally. One popular choice of sensors includes a Laser Measurement System (LMS) and a vision system. Good examples of robots using LMS and vision are vehicles entering the annual Intelligent Ground Vehicle Competition (IGVC) and competing in the 2005 Grand Challenge sponsored by the Defense Advanced Research Projects Agency (DARPA). This paper focuses on one method of integrating non-stereoscopic vision (camcorder) information with laser distance measurements. First, background information on one such AGRV mechanical structure and a sensor suite is provided. This platform allows testing of algorithms using real hardware. The paper also explains the AGRV processes and image management. The core presentation concerns the method used for integrating LMS with vision. Once integrated, LMS and vision act as one set of data with one format, yet the method exploits all the information available from both. Finally, the paper illustrates one way to use this processed information for finding paths through a field of obstacles and road edges. AGRV PLATFORMThe Center for Applied Research and Technology (CART) at Bluefield State College designed and built an AGRV the students called "V2". Being a little larger than an electric wheelchair and weighing slightly less than 300 pounds, the vehicle has a control system that gives the robot superb maneuverability. A full suite of sensors allow the robot to sense many aspects about its environment. The particular sensor suite for the AGRV allows algorithms to mimic human decision making. Therefore, our vehicle provides an excellent platform for studying various autonomous algorithms such as the ones presented in this paper. This section will present the hardware design of for the vehicle in three parts: the mechanical system, the electrical system, and other design concerns. Mechanical SystemThe overall mechanical design focuses on simplicity, durability, compactness, maintainability, and most importantly, safety. The vehicle is designed to operate and navigate safely in both indoor and outdoor environments. This small and versatile design provides the opportunity to test and develop the human-like system on a fully functional platform. The mechanical design can be divided into three separate categories:Page 10.52. Vehicle FrameThe vehicle frame is constructed of steel tubing. Steel tubing was chosen due to its light weight, durability, and ability to house wiring. The tubing acts as a conduit to conceal and organize connections as well as to shield vulnerable lines from RF noise. The rectangular design allows the frame to be strong while creating a protective carriage that houses the batteries, chargers, and other various components. Drive SystemOur ARGV u...

No abstract

No abstract

This paper presents the design and provides a partial analysis of the performance of an autonomous ground robotic vehicle called Vasilius. Applications for Vasilius include autonomous navigation on a somewhat marked path with obstacles, leader following, and waypoint navigation. The paper focuses on three aspects of Vasilius: the design, the performance, and a technique for filtering, mapping, and learning. The design of Vasilius embodies a novel idea of modeling an autonomous vehicle after human senses and the human decision-making process. For instance, Vasilius integrates information from seven types of independent sensors, and categorizes them into either short-range reaction sensors and/or long-range planning sensors, analogous to what the human brain does. The paper also analyzes the performance of Vasilius, relating theoretical predictions to actual behavior. Some of these analyses, especially for the filtering, mapping, and learning, are still in progress. Performance measures that have been measured include speed, ramp climbing, turn reaction time, battery life, stop reaction time, object detection, and waypoint accuracy. Finally, the paper discusses Vasilius' use of a new approach to filtering, mapping, and learning to enhance its performance.