A STUDY TO DETERMINE THE FEASIBILITY OF DEVELOPMENT OF AN AREA TRAFFIC SURVEILLANCE AND CONTROL SYSTEM

A TWENTY-TWO MONTH STUDY WAS CONDUCTED TO DETERMINE THE FEASIBILITY OF DEVELOPING AN AREA TRAFFIC SURVEILLANCE SYSTEM FOR COLLECTION OF TRAFFIC DATA FOR USE IN RESEARCH DEALING WITH THE THEORY OF TRAFFIC FLOW, AND FOR APPLICATION TO REAL TIME CONTROL OF A SYSTEM OF STREETS AND HIGHWAYS. DURING THE STUDY A SURVEY OF EXISTING SURVEILLANCE SYSTEMS WAS PERFORMED AND THE CAPABILITIES OF THESE SYSTEMS WERE EVALUATED. UPON COMPLETION OF THIS REVIEW, NEW AND ADVANCED DETECTION SYSTEMS AND MOUNTING PLATFORMS WERE INVESTIGATED AND SEVERAL SURVEILLANCE SYSTEMS WERE FORMULATED AND EVALUATED ACCORDING TO NEWLY DEVELOPED COST-EFFECTIVENESS CRITERIA. RECOMMENDATIONS WERE MADE CONCERNING THE FEASIBILITY OF EACH SYSTEM, THE AREAS OF ADDITIONAL RESEARCH WERE IDENTIFIED, AND NECESSARY DEVELOPMENT AND DEMONSTRATION PROGRAMS WERE OUTLINED. THE RESULTS OF THE STUDY SHOW THAT DIFFERENT TYPES OF SURVEILLANCE SYSTEMS ARE NEEDED TO SATISFY THE REQUIREMENTS OF A REAL TIME CONTROL SYSTEM AND OF A SYSTEM FOR COLLECTION OF TRAFFIC FLOW DATA FOR RESEARCH PURPOSES. A GROUND-BASED SYSTEM USING DISCRETE SENSORS AND HARDWIRE TELEPHONE COMMUNICATION CHANNELS IS THE MOST ATTRACTIVE SURVEILLANCE DEVICE FROM THE POINT OF VIEW OF COST AS WELL AS OF RELIABILITY. THE DISCRETE SENSOR SYSTEM CAN ALSO BE USED FOR COLLECTION OF RESEARCH DATA, BUT THE AREA UNDER SURVEILLANCE WOULD HAVE TO BE SATURATED WITH SENSORS WHICH WOULD SUBSTANTIALLY INCREASE ITS COSTS. A MM RADAR DEVICE APPEARS TO HAVE PROMISE FOR LOCAL AREA COVERAGE IN A REAL TIME SURVEILLANCE SYSTEM. HOWEVER, THE COST IS GREATER THAN THAT OF THE DISCRETE SENSOR, AND A CONSIDERABLE AMOUNT OF HARDWARE DEVELOPMENT WORK WOULD BE NEEDED TO PERFECT SUCH A RADAR SENSOR. THE RADIO FREQUENCY LOOP DETECTOR WAS FOUND TO BE THE MOST PROMISING AND RELIABLE SENSOR FOR THE CONTROL APPLICATION. FOR COLLECTION OF RESEARCH DATA, LARGE VIEW SYSTEMS WERE FOUND TO BE MOST PROMISING. AERIAL PHOTOGRAPHY IS CAPABLE OF PRODUCING THE REQUIRED MEASUREMENTS OF TRAFFIC FLOW TO THE DESIRED ACCURACY, BUT THESE SYSTEMS ARE VERY SENSITIVE TO INCLEMENT WEATHER CONDITIONS AND TO POOR VISIBILITY. THE MOBILITY OF AIRBORNE SYSTEMS, ON THE OTHER HAND, MAKE THEM IDEALLY SUITED FOR USE AS RESEARCH TOOLS. A PROBLEM OF GREAT SIGNIFICANCE IN ALL LARGE VIEW SYSTEMS IS THAT OF DATA REDUCTION. THE STUDY DESCRIBES METHODS FOR AUTOMATIC REDUCTION OF DATA FROM AERIAL PHOTOGRAPHS, AND OUTLINES A PROGRAM OF FUTURE RESEARCH IN THIS AREA. /BPR/