Advanced SSTK surveillance analytics integrates, analyzes, and correlates real time data from radar, cameras, motion detectors, and ground sensors, and other physical surveillance sensors to amplify and pinpoint threat detection, simplifying surveillance of complex environments. SSTK combines data from multiple sensors to detect moving objects of significance while ignoring motion due to changes in lighting and weather conditions. The analytics identifies movement patterns to classify objects by type such as people, automobiles, animals, or debris, and assess potential threats.
Identifying and correlating related data from different sources takes SSTK beyond common video analytics into full sensor suite analysis that quickly identifies and more accurately classifies threats with fewer false identifications. The SSTK technology:
- Analyzes movement patterns to identify potential threats
- Provides multiple looks at each target
- Takes advantage of the strengths of each sensor type to increase target identification and location accuracy.
- Handles conflicting data
The technology is based on advanced multiple hypothesis, Gaussian sum and Non-Gaussian tracking, non-Gaussian registration, and Bayesian inference techniques to significantly improve search and surveillance. These analytical techniques combine data from a wide array of sensors into a single, unambiguous data set that resolves conflicting data, continuously updates earlier results, and forms the basis for all other SSTK functions and operations.
Non-Gaussian Target Tracking Employing Monte Carlo methods
Tracking a target’s path through a network of different types of sensors can be a complex process, particularly in bad weather, at night, or when multiple targets and other moving objects are present. The SSTK target tracking method is based on an advanced mathematical method that derives path information by predicting many options for the target’s future route and honing in on the path best supported by sequential sensor detections and non-detections. Target tactics and other sophisticated types of non-linear motion are modeled internally. Model precision is continuously improved in real-time based on sensor contacts using a mathematical method developed by Daniel Wagner for Naval operations analysis. The result is an accurate and rapidly calculated track based on information from all types of sensors in real-time.
Key Advantages of Non-Gaussian Fusion
- More accurately determines which sensor contacts are associated with which targets
- Provides better estimates of the position and velocity of targets of interest
- Allows the generation of optimal resource allocation plans
- Provides much more accurate modeling of line-of-bearing data, including the use of signal propagation information
- Allows for the use of negative information from sensors that are not detecting the target, which in many cases can be nearly as valuable as positive information when estimating target position
- Allows for the accurate modeling of complex target motion and tactics such as a target attempting to avoid active prosecution.