Integration and Control of Reactive Visual Processes

This paper presents a software skeleton system for experiments in integration and control of real time active vision systems. This skeleton has been constructed by a consortium of six laboratories as part of a long term basic research investigation of issues of integration and control of real time vision. The first chapter describes the problems of integration and control and the context for their investigation The second chapter presents the SAVA III skeleton system which has been developed to support our experiments in integration and control. An overview of the skeleton system is presented, and then the communication mechanisms are described. The use of a rule based interpreter based on CLIPS 5.1 permits each module to send messages which can define rules and functions within the other modules, essentially allowing modules to program each other. The rule interpreter is used to construct the control component for the processing cycle for each module. The interpreter is also used for the control part of pre-attentive demons which detect perceptual events within the modules. The SAVA system is illustrated with a description of a recent demonstration of a visual navigation system constructed in the SAVA skeleton. The components of the visual navigation system are presented and details are given concerning the use of stereo for obstacle detection, and the use of landmarks for updating the estimated position.