A simple approach to a vision-guided unmanned vehicle

2011 14th International IEEE Conference on Intelligent Transportation Systems (ITSC), 2011

As interest on autonomous vehicles is growing worldwide, different approaches, based on different perception technologies and concepts, are being followed. This paper exposes the importance of the use of vision technology in most of these approaches, and presents the experience of the SNUCLE autonomous vehicle which successfully completed the Hyundai Autonomous Challenge in November 2010.

Control Engineering Practice, 1996

The development of autonomous visual guidance for land vehicles at UniBwM is reviewed. An initial breakthrough in performance level was achieved in 1986 by combined differential/integral representations of the road skeleton and the vehicle state in conjunction with recursive estimation techniques (4-D approach). Feedback control laws allowed easy implementation of reactive behavior; by complementing this approach with generic feedforward control-time histories for several mission elements, a capability for mission performance is introduced. Lane changes, turn-offs and handling forks in the road are the most essential such elements; in conjunction with a capability for landmark recognition and digital map reading this allows the autonomous performance of entire missions. Recently developed capabilities of the VaMoRs test vehicle are discussed.

2001

documents a National Institute of Standards and Technology (NIST) Intelligent Systems Division (ISD) effort to determine the status of Automatic Guided Vehicles (AGVs) in the United States. This report is intended to: 1) provide information to management for guiding future ISD and MEL projects and 2) review ISD mobility project research that was focused mainly on the industrial vehicle industry. The particular information collected parallels NIST's main mission goals, standards and measurements, as well as ISD advanced technology strengths: sensing, modeling, and planning for autonomous vehicle navigation. This work falls under the Intelligent Control of Mobility Systems program. Details of the program can be found at . The methods for determining the state of the AGV industry's measurement, standards and technology needs were as follows: 1) Search Internet, read trade-magazines, attend trade shows, and visit AGV vendors and users. 2) Develop initial list of AGV manufacturers and visit key manufacturers. Determine size, capabilities and plant locations. Gather as much information as possible regarding current state-of-the-art in sensing, modeling, planning and navigation. 3) Develop list of AGV end-users and visit key-users, particularly government users. Determine scenarios, needs, limitations and desires of end-users. Vehicle intelligence has been advancing for many years, mainly in sensing and navigation. Researchers have developed many key advancements in vehicle navigation including this small subset: • J. Evans, B. Krishnamurthy, Helpmate Robotics, Inc. [4]-developed an autonomous mobile robot courier for material transport in hospitals and used around the world operating in uncontrolled and unsupervised environments up to 24 hours per day. Navigation via sonar, vision and other sensors was used. • H. Schneiderman, M. Nashman, NIST [11] -demonstrated vision-based, autonomous, road-following of a HMMWV (high mobility, multi-wheeled vehicle) on highways and rural roads at full speeds using a real-time, image-processing algorithm reading lane-markings and confined only to the 2-D plane. • D. Raviv, Florida Atlantic University and M. Herman, NIST [9]-demonstrated vision-based autonomous road following in the laboratory using optical-flow-based theory and generating motion commands directly from a visual feature or cue, consisting of the projection into the image of the tangent point on the edge of the road, along with the optical flow of this point. This approach requires no 3-D scene reconstruction and uses relatively simple computations. • J. Albus, NIST, 4-D/RCS Reference Model Architecture for Unmanned Ground Vehicles, [1]-currently developing a reference model architecture for the Demo III Experimental Unmanned Vehicle program where 4-D/RCS (4 dimensional/real-time control system) integrates the NIST RCS with the German (Universitat der Bundeswehr Munchen) VaMoRs 4-D approach to dynamic machine vision, including computational nodes containing behavior generation, world modeling, sensory processing, and value judgment processes. • K. Murphy, et. al., NIST [8]-demonstrated a HMMWV (high mobility, multi-wheeled vehicle) with sensors and computer controlled actuators that detects, using a 30 x 60 pixel field of view laser range scanner, and avoids obstacles while it drives off road at speeds up to 35 km/h. The planner computes smooth, obstacle free paths that follow an operators commanded path. • X. Deng, E. Milios, et. al. York University [3]-demonstrated landmark selection for minimal sensing cost (detection and tracking) and a strategy that verifies a map in edge traversals only, using a single edge marker, that solves the issue of cumulative odometry errors which may cause a robot to lose track of its position in the real world.

FIE'99 Frontiers in Education. 29th Annual Frontiers in Education Conference. Designing the Future of Science and Engineering Education. Conference Proceedings (IEEE Cat. No.99CH37011

In this paper we relate our experience with the integration of the International Ground Robotics Competition (IGRC) [1] into the Electrical and Computer Engineering senior capstone design course sequence at the University of Detroit Mercy. The use of a competition-driven Autonomous Ground Vehicle (AGV) design for a senior capstone program offers a number of benefits and challenges [2],[3]. The attractive features include: the student excitement and motivation created by the competitive environment; the fact that an AGV design is inherently multidisciplinary in nature involving mechanical and electrical system integration, power electronics, feedback control, digital electronics, software development, sensors, safety, power management etc.; the competition-driven format mimics the need in industry to carry out deadline-conscious product design; and the sophistication of subsystems like vision, obstacle detection, path planning, and feedback control offer good opportunities for graduate and undergraduate research projects. The challenges include: maintaining focus on proper design methodology and avoiding ad hoc ends justify the means design [4]; the extensive fund raising necessary to pay for the expensive component systems; the substantial faculty time and energy necessary to support such a comprehensive design effort; and the course structural organization required to support a contest deadline occurring after the first five weeks of the second semester of the two semester capstone design course sequence.

IEEE Journal on Robotics and Automation, 1987

A modular system architecture has been developed to support visual navigation by an autonomous land vehicle. The system consists of vision modules performing image processing, three-dimensional shape recovery, and geometric reasoning, as well as modules for planning, navigating, and piloting. The system runs in two distinct modes, bootstrap and feedforward. The bootstrap mode requires analysis of entire images to find and model the objects of interest in the scene (e.g., roads). In the feedforward mode (while the vehicle is moving), attention is focused on small parts of the visual field as determined by prior views of the scene, to continue to track and model the objects of interest. General navigational tasks are decomposed into three categories, all of which contribute to planning a vehicle path. They are called long-, intermediate-, and short-range navigation, reflecting the scale to which they apply. The system has been implemented as a set of concurrent communicating modules and used to drive a camera (carried by a robot arm) over a scale model road network on a terrain board. A large subset of the system has been reimplemented on a VICOM image processor and has driven the DARPA Autonomous Land Vehicle (ALV) at Martin Marietta's test site in Denver, CO.

International Journal for Research in Applied Science & Engineering Technology, 2021

A significant perspective related with vehicles is their speed. A faster vehicle encourages us to arrive at our destination in lesser time, sparing our valuable time. In any case, tragically, we have been seeing the ascent in vehicular mishaps because of uncontrolled speeding by the drivers. The traffic signs alone cannot ensure the safety of vehicles since it is dependent upon the driver to adhere to the directions. Likewise, there is the situation of human-mistake where the driver essentially neglects to pay special mind to the traffic signs. This paper focuses on the road traffic sign detection systems which help in informing the intelligent vehicle about the possible road conditions ahead and be cautious about it with the help of Image processing. A module which consists of a Raspberry Pi or USB camera with a wide view and a simple processor is installed on the vehicle. The developed system works with three different stages: image pre-processing, detection, and recognition. The entire developed system is programmed using Python incorporated with OpenCV library and implemented using open source hardware platform and open-source software environment.

2019 Innovations in Power and Advanced Computing Technologies (i-PACT), 2019

In this paper, the hardware setup of an autonomous robotic vehicle is developed. The controller of the vehicle is developed based on the computer vision technology which became more smart due to the application of popular tool open CV. An algorithm based on color detection is developed to navigate the vehicle in different directions. Raspberry Pi camera is used to access images which are the indications of the directions of the movements such as stop, left turn, right turn. The efficacy of the proposed algorithm is verified experimentally.

2006

This paper describes the software architecture of Stanley, an autonomous land vehicle developed for high-speed desert driving without human intervention. The vehicle recently won the DARPA Grand Challenge, a major robotics competition. The article describes the software architecture of the robot, which relied pervasively on state-of-the-art AI technologies, such as machine learning and probabilistic reasoning. * The Stanford Racing Team gratefully acknowledges the support of its sponsors,

Society of Photo- …, 2010

In this paper, we present Argos, an autonomous ground robot built for the 2009 Intelligent Ground Vehicle Competition (IGVC). Discussed are the significant improvements over its predecessor from the 2008 IGVC, Kratos. We continue to use stereo vision techniques to generate a cost map of the environment around the robot. Lane detection is improved through the use of color filters that are robust to changing lighting conditions. The addition of a single-axis gyroscope to the sensor suite allows accurate measurement of ...

Journal of Field Robotics, 2004

Black Knight, the University of Central Florida's vehicle in the 11th Intelligent Ground Vehicle Competition (IGVC) competed in 2003. Completing in 5th place in the navigational challenge and 10th in the autonomous challenge in its first competition has proven our vehicle to be a strong competitor in this competition. The vehicle has many interesting features that allow it to achieve its success. The vehicle's 300 lb. capacity allows for two onboard full-sized computers and two 12 V marine batteries that power the computers for up to 2 h. The vision system is not a simple reactive system but rather it classifies its view into objects and builds a map of the territory as it learns of its features while traveling. Two transformations and the location data from the GPS and other sensors are used to associate the locations in the image to locations in the map. The operations of the vehicle are modeled after the typical operations of a ship. We have programs that perform the functions of the captain, the helm, the navigator, and the engineer. In addition we have a program performing sensor data fusion from the GPS, compass, and wheel encoder data. The navigation uses an adapted two-dimensional approximate cell decomposition method that satisfies the nonholononic constraints of our vehicle and allows it to find the shortest path to the goal while avoiding all obstacles.