AN INTEGRATED LASER RANGER AND CAMERA SYSTEM

PRESENT ROBOT VISION SYSTEMS REQUIRE EITHER MULTIPLE CAMERAS OR A LASER RANGING/SCANNING SYSTEM TO ACQUIRE VISUAL INFORMATION FOR CHARACTERIZING SCENES IN THREE DIMENSIONS. THE HUMAN EYE WITH ITS ASSOCIATED INTELLIGENCE SYSTEMS DOES A REMARKABLE JOB OF CONSTRUCTING A THREE DIMENSIONAL IMAGE WITHOUT USING A RANGE SENSOR. ATTEMPTS TO EMULATE THE PERFORMANCE OF THE EYE WITH CCD CAMERAS AND MASSIVE COMPUTING EFFORTS HAVE NOT BEEN TOO SUCCESSFUL. LASER RANGING/SCANNING SYSTEMS ALSO HAVE THEIR LIMITATIONS IN THAT THEY ARE CURRENTLY MECHANICALLY SCANNED, LOW RESOLUTION AND REQUIRE A SIGNIFICANT AMOUNT OF PROCESSING TO PRODUCE AN IMAGE. THE OBJECT OF THIS RESEARCH IS TO INVESTIGATE THE FEASIBILITY OF MERGING THE DATA FROM A CCD CAMERA OR A FOCAL PLANE PROCESSOR WITH THE RANGE AND REFLECTANCE DATA FROM A LASER RANGER. PHASE I WILL CONSIST OF DEVELOPING THE ALGORITHMS AND EXPERT RULES FOR FUSING LASER AND CAMERA DATA AND DEMONSTRATING WITH A COMPUTER SIMULATION. IT WILL ALSO INCLUDE A DESIGN APPROACH FOR IMPLEMENTING A CAMERA INTEGRATED WITH LASER RANGER. WE ANTICIPATE THAT THE PHASE II EFFORT WILL RESULT IN A PROTOTYPE LASER RANGING CAMERA SYSTEM CONTROLLED BY A RULE BASED SYSTEM.