Adaptive control of cooperating sensors - focus and stereo ranging with an agile camera system

This paper presents a cooperative computer vision procedure in which focus ranging and stereo ranging operate together, verifying the results of each in computing the position (but not the shape) of arbitrary objects in a stationary, unknown environment. The procedure increases the reliability of the position measurements by enforcing measurement consistency via mutual constraint, and increases their accuracy by combining them with a maximum likelihood estimator into an estimate of lower variance than any of the measurements alone. The final outcome of the procedure is a set of estimated three-dimensional points together with their estimated uncertainties built from a sequence of dynamic, adaptive sensing operations. The results of 75 experiments processing close to 3000 different obiect points lying between 1 and 3 meters distant from the cameras show that the integrated range values are (I) highly reliable, since no mistaken combined range measurements are observed, and (II) more pre cise than either of the computed ranges alone. The cooperative methodology extends to more and different sensors, and the results lend practical credence to the view that multiple sensors (I) allow the limitations of a single sensor to be circumvented, (II) provide larger, statistically more effective data sets, (III) reduce mistakes generated by inaccurate interpretation models, using sensors outside of their known operating regions, and sensor failure.