Norman C. Griswold

Visual control of an autonomous vehicle (BART)-the vehicle-following problem

The authors consider the problem of vehicle-following including automatic steering and speed control of an autonomous vehicle following the motion of a lead vehicle. A visual control system for vehicle-following is presented. The system consists of the following modules: image processing, recursive filtering, and a driving command generator. First, the range and heading signal of the lead vehicle are obtained by visually identifying a unique tracking feature on the lead vehicle. Based upon this information, appropriate steering wheel and speed commands for driving are generated, which are then downloaded and executed on a microprocessor controller. The visual control system was tested on BART (Binocular Autonomous Research Team), a testbed vehicle developed at Texas A&M University for autonomous mobility. Successful full-scale test runs have been accomplished for speeds up to 20 mi/h. >

Stop-sign recognition based on color/shape processing

This paper presents a robust vision-based stop-sign reconition technique based on sequential processing of color and shape. The primary red-green-blue color coordinate system is first transformed into the saturation-hue-brightness color coordinate system. This color coordinate system allows the red color area of a stop sign to be bounded under various brightness conditions caused by weather, sun angle, or shadows. A combination of a median filter, a morphological filter, Sobel edge operator, and Hough transform is then employed to obtain the boundary contour. It is demonstrated that the parameters of eight straight lines representing the octagonal sides are sufficient for this purpose. Experimental results indicate that stop signs are successfully distinguished from other traffic sighs and background clutter.

A generalized block truncation coding algorithm for image compression

Block Truncation Coding is a recently developed approach to image compression whose design is specified by the appropriate moment preserving quantizer. In this paper we show how the basic Block Truncation Coding algorithm can be generalized to include a family of moment preserving quantizers with the potential for improved performance. We then illustrate by way of example that such improvement is indeed possible from the standpoint of peak signal to noise ratio. There is a subclass of this family of moment preserving quantizers for which practical difficulties in implementation exist; however, we show that frequently we can avoid this subclass and still obtain good performance.

Control for mobile robots in the presence of moving objects

The problem of collision-free navigation and guidance for mobile robots is essential to the survival of unmanned terrain vehicles. This communication adds complexity to the guidance problem by introducing moving objects that may cross the desired navigational path of an autonomous vehicle. The vehicle is equipped with sensors that report locations of the moving objects. Speed and direction of each moving object are modeled as random walk processes. An optimization approach that provides for the acceleration or deceleration of the vehicle is developed, as is the duration of the time for which this control must be applied to avoid a collision with moving objects. The constraints on the objective function are given in terms of collision probabilities, and simulation results are presented for a collision-free environment. Implications of this approach for autonomous vehicles are given in terms of the flexibility of path planning. >

An invariant traffic sign recognition system based on sequential color processing and geometrical transformation

One of the most noteworthy problems associated with conventional pattern recognition methods is that it is not easy to extract feature vectors from images which are not translation, rotation, and scale change invariant in outdoor noisy environments. This paper describes the development of an invariant traffic sign recognition system capable of tolerating the above variations. The signs are restricted to three types of warning signs and are all of red color. The developed method is insensitive to brightness changes as well as invariant to translation, rotation, scale change, and noise. The architecture of this system is based upon neural network supervised learning after geometrical transformations have been applied. The performance of this system is compared with other invariant approaches in terms of the percentage of correct decisions in outdoor noisy environments.<<ETX>>

Adaptive IIR filter design for single sensor applications

The goal of this research was to investigate the theoretical design and physical implementation of a digital adaptive IIR filter to serve as an enhancement to the traditional active RC or passive RLC anti-aliasing filter. This all-digital filter will reside directly on the DSP engine. As explained in the paper, the adaptive IIR filter is designed to process an oversampled signal coming from a single sensor to reject noise in an acquisition system. Differentiation between the noise and the signal is obtained by exploiting the different auto-correlation functions of the two signals. In contrast to oversampling techniques employed in sampled data systems that are designed to relax the requirements of an analog anti-aliasing filter, this filter will track a signal in the frequency domain. Several power spectral density plots are given to illustrate the performance of the new filter. The results also indicate that the new filter performs well as compared to the Wiener filter in the stationary case.

Perceptual Coding In The Cosine Transform Domain

Source encoding for digital image transmission is revisited with an energy distribution approach in the perceptual domain. Past investigations have utilized power spectral density in conjunction with the Frei eye model and full image Fourier transform coding. In this investigation, the cosine transform is utilized on a partitioned image. A cosine energy function is defined and weighted by the eye model. This results in a circular symmetric form of a bit map which simplifies source coding. This approach outperforms a standard bit allocation procedure allowing graceful degradation at 1, .75, and .5 bits/pixel. Analysis includes the perceptual mean square error and peak signal-to-noise ratio as metrics of performance. This procedure suggests a more rapid hardware implementation.

A new stereo vision model based upon the Binocular fusion concept

Abstract In computer vision, the idea of using stereo cameras for depth perception has been motivated by the fact that in human vision one percept can arise from two retinal images as a result of the process called “fusion”. Nevertheless, most of the stereo algorithms are generally concerned with finding a solution to obtaining depth and three-dimensional shape irrespective of its relevance to the human system. Recent progress in the study of the brain mechanisms of vision has opened new vistas in computer vision research. This paper investigates this knowledge base and its applicability to improving the technique of computer stereo vision. In this regard, (1) a stereo vision model in conjunction with evidences from neurophysiology of the human binocular system is established herein; (2) a computationally efficient algorithm to implement this model is developed. This algorithm has been tested on both computer generated and real scene images. The results from all directional subimages are combined to obtain a complete description of the target surface from disparity measurements.

A note on adaptive block truncation coding for image processing

In this correspondence, we modify a recent approach for image compression, block truncation coding, to impart block-by-block adaptivity. We then illustrate via example images that improved performance is obtained.

A transportable neural network controller for autonomous vehicle following

This paper presents the development and testing of a neural network controller for autonomous vehicle following. Autonomous vehicle following is defined as a vehicle controlling its own steering and speed while following a lead vehicle. The strength of the developed controller is that no characterization of vehicle dynamics is needed. As a result it can be transported to any vehicle regardless of its nonlinear and often unobservable dynamics. Data for the range and heading angle of the lead vehicle were collected for various paths with a human driver performing the vehicle following control function. The data was collected for different driving maneuvers including straight paths, lane changing and right/left turns. Two time-delay backpropagation neural networks were then trained based on the data obtained under manual control, one network for speed control and the other for steering control. After training, the vehicle following was done using the trained neural network controller. The results obtained and presented on a video tape indicate that it is feasible to employ a neural network to satisfactorily perform autonomous vehicle following.