Estimating density and number of lane change maneuvers on a freeway

This paper illustrates a methodology for estimating density in a freeway lane between detector stations and measuring the net number of vehicles to enter (or leave) the lane, i.e., the lane inflow. The work uses vehicle arrivals at each detector station and information from a sparse vehicle reidentification system, one that may match the measurements from as few as five percent of the vehicles that pass both detector stations. Such reidentification systems have already been documented and deployed for dual loop detectors and they will be used for illustration. Of course this work is also applicable to other vehicle reidentification systems, which may be based on other detection technologies, that may have a higher rate of reidentification.