Multiple sensor tracking contains a fundamental problem: determining which tracks from various sensors originated from the same source. The techniques for this process vary depending on types of sensors. If all sensors are active, then each one can create a 3D track. A 3D distance metric can then be used to associate the individual sensor tracks with each other. But if all sensors are passive, the approach is different and can be ad hoc. Aside from a scissor or stereo angle which can distinguish one of the two passive dimensions, ad hoc ways to correlate passive sensor tracks include assuming altitudes associated with time of track to hypothesize a 3D position for correlation. A single algorithm correlating multiple passive and active sensors (where all the sensors can be passive; all active; or a combination) was presented in 2015. That paper produced an algorithm which computes, in closed form, a weighted least squares best fit general polynomial to the asynchronous measurements from multiple sensors, whether they be active, passive or a combination. The score of the best fit polynomial to the measurements can be used to determine which track combinations should be correlated via an assignment algorithm. This algorithm uses the angles-only (for passive sensors) and angles plus range (for active sensors). This single algorithm for the various combinations of sensors uses no ad hoc methods. However, as this paper will demonstrate, the algorithm can be taken even farther. What if some of the sensors are passive with long integration times and could measure angle rate? What if some of the active sensors could measure range rate? This paper extends the results presented in 2015 to sensors that also measure angle rate and range rate. Because this single algorithm is closed-form and fast, it can be used at the heart of a multiple sensor correlator and can now employ all elements of each sensor’s measurement: (1) angles only; (2) angles and angle rate; (3) angles and range; and/or (4) angles, range and range rate. No longer is there a need for specialized logic and several algorithms to handle various combinations of sensors.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Closed Form Multiple Disparate Sensor Correlation/Fusion


    Beteiligte:
    Roecker, J.A. (Autor:in)


    Erscheinungsdatum :

    02.03.2024


    Format / Umfang :

    3518837 byte




    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Disparate sensor real-time scene fusion for tactical environments

    Selleck, Matthew B. / Burke, David / Johnston, Chase | British Library Conference Proceedings | 2019


    Target Location and Identity Estimation and Fusion using Disparate Sensor Data

    Naidu, Parthsarathy / Mudassar, Mohammed / Girija, G et al. | AIAA | 2005


    Measurement and Track Fusion with Disparate Sensors

    Dunham, Darin T. / Vandiver, Robert C. / Zutshi, Arjun D. | IEEE | 2024