Road-map assisted ground moving target tracking

Abstract

Tracking ground targets with airborne GMTI (ground moving target indicator) sensor measurements proves to be a challenging task due to high target density, high clutter, and low visibility.The exploitation of nonstandard background information such as road maps and terrain information is therefore highly desirable for the enhancement of track quality and track continuity. The present paper presents a Bayesian approach to incorporate such information consistently. It is particularly suited to deal with winding roads and networks of roads. The target dynamics is modeled in quasi one-dimensional road coordinates and mapped onto ground coordinates using linear road segments taking road map errors into account. The case of several intersecting roads with different characteristics, such as mean curvature, slope, or visibility, is treated within an interacting multiple model (IMM) scheme. Targets can be masked both by the clutter notch of the sensor and by terrain obstacles. Both effects are modeled using a sensor-target state dependent detection probability. The iterative filter equations are formulated within a framework of Gaussian sum approximations on the one hand and a particle filter approach on the other hand. Simulation results for single targets taken from a realistic ground scenario show strongly reduced target location errors compared with the case of neglecting road-map information. By modeling the clutter notch of the GMTI sensor, early detection of stopping targets is demonstrated.

Efficient planning under uncertainty for a target-tracking micro-aerial vehicle

“…In cases of multiple-hypothesis tracking of a single target, the belief updating is traditionally done with particle filters ( [4], [5], [6]). An exception is [19], which explores both the Gaussian sum approximation and particle filter approaches. However, these algorithms typically focus on the problem of performing accurate belief updating and data association, rather than on the decision-making for the agent.…”

Airborne behaviour monitoring using Gaussian processes with map information

“…Specific applications under consideration include, but are not limited: border patrol, search and rescue, surveillance, and law enforcement. In particular, UAV surveillance with an onboard GMTI (ground moving target indicator) radar or SAR (synthetic aperture radar) can provide a certain level of accurate estimation of massive sets of targets and consequently allow detection of possible threats [1], [2], [3]. However, the operators still need to analyse the gathered data and construct a picture of events to detect suspicious behaviours.…”

Integrating Spline Curves in Road Constraint Object Tracking

“…For ground moving objects assumptions on road-or track-constrained motion may augment significant information to tracking processes. To obtain robust object tracking results, there exist several concepts to include road map information into the tracking algorithm [2] [3].…”

“…straights, transition curves and circular arcs, to enable a comfortable and save driving without any abrupt variations in lateral acceleration. A prominent way to store this geometric knowledge are piecewise defined linear polynomial models [3] [4]. If additional information about road tangents is available a more precise geometric approximation can be achieved, making use of higher order polynomials.…”

“…The second class of concepts formulates the dynamic behaviour in local curve coordinates [3] [6]. If the representation of the road network fulfils several preconditions this strategy yields in robust results.…”