This paper discusses problems in image processing, pattern recognition and control associated with the guidance and control of automated surface vehicles. An overview of work done by others is presented first. This is followed by a discussion on obstacle detection and avoidance methods and a description of existing and specially developed algorithms to be applied to guidance. A brief description is presented of the model to be used for the discrete control of the dynamic system. It is needed to determine such things as the optimal control law and most important, it allows calculation of the necessary sampling rate for stability which is the time available for real time computation between samples. Navigation is a complicated problem which may make use of stored information about the highway. Guidance and safety are problems which must be solved in real time under widely variable weather, seasonal and road conditions. Its complete solution will require a relatively long time, but some useful results may be obtained in the near future.
[1]
Stephen D. Shapiro,et al.
Feature space transforms for curve detection
,
1978,
Pattern Recognition.
[2]
Takeshi Hirose,et al.
An Automobile with Artificial Intelligence
,
1979,
IJCAI.
[3]
R. J. Klensch,et al.
Harmonic radar helps autos avoid collisions
,
1973,
IEEE Spectrum.
[4]
Robert E. Fenton,et al.
On Vehicle Automatic Longitudinal Control
,
1977
.
[5]
R.E. Fenton,et al.
On the steering of automated vehicles—A velocity-adaptive controller
,
1980,
IEEE Transactions on Vehicular Technology.