GeoEvent can alert you when two features intersect or are near each other using several methods including built-in filters and the Incident Detector Processor. However, running this analysis against time could be more difficult depending on the scope of your project. Can you explain more regarding the amount of time needed for each intersection as well as the specific application of this potential GeoEvent Service?
The Track Gap Processor within GeoEvent has a built-in temporal element that tracks a time interval when events are not received (think a car sitting idle and not changing positions for a given amount of time). We could attempt to configure this in reverse to fit the scope of your project (to alert when GeoEvents are received for a given time interval). Combining this with a spatial filter will ensure that the points are near each other to trigger the processor to start tracking.
Thanks Chris for the reply!
I have a rather complex working GEP topology at the moment that uses filters and the incident detector. I could not see a solution for configuring an incident detector or filter when a point was within X nautical miles from a polygon or line, as I did not see a radius parameter. I looked at the ConvexHull and Envelope Creator processors, but those did not have a radius parameter.
The use case is that if a entity (point geometry) is 0 to X NM outside of a Geofence (which could be a line or poly), generate an incident.
If there is a solution specially configured single incident detector that could process this, I would appreciate knowing this. The only solution I could think of for this case was to use a filter and a fence - the primary geofence and a larger filter that represented the radius around the fence I am trying to generate the detection of.
I don't think that Track Gap Processor could help. A Point track sitting idle or not changing positions, would still be reporting its position. A track Gap would only occur if the track stopped reporting (I think).
An ideal solution would be to have the IncidentDetector processor open sourced so users could simply extend it. This would allow users to create custom incident detectors and create simpler GEP topologies.
Does the NM radius fence around the primary geofence change? For example, can we assume that this secondary radius is always 2NM wide? I've got a few more conceptual ideas that I'd like to try on my own, and for these tests I'll be assuming that each point can be between 0-2NM from the primary geofence. Currently all processors can be modified by administrators by leveraging the GeoEvent SDK that is packaged with every installation. The SDK contains a Developer Guide and SDK reference for anyone interested in modifying the components. This does however require a knowledge of the code needed to make the modifications, and currently there is no intuitive builder or interface to help with non-code modifications.