Maneuvering target tracking method under constraint conditions

Abstract

The invention discloses a maneuvering target tracking method under constraint conditions, comprising the steps of: A, obtaining the position measurement information of a maneuvering target; B, converting the position measurement information into Cartesian coordinate measurement; C, establishing a constrained coordinate static model in a one-dimensional space, i.e., building the mileage S expression of the position of the maneuvering target in road x, y directions; D, setting motion models according to motion characteristics of the maneuvering target, and establishing a state equation and a measurement equation according to the established expression; and E, for each motion model, filtering converted Cartesian coordinate measurement to obtain a one-dimensional variable mileage S and speed filtering result. According to the embodiment, the tracking performance of a maneuvering target can be improved.

Description

technical field [0001] The invention relates to maneuvering target tracking under constraint conditions, in particular to a maneuvering target tracking method for transferring two-dimensional constraints to one-dimensional space for processing. Background technique [0002] In the field of radar target tracking, there are some scenarios where the development of the target's trajectory is not determined by its own movement speed, but is seriously affected or even strictly restricted by the external environment. For example, a vehicle driving on a road is constrained by terrain and road shape, and the possibility of leaving the road is extremely small, which is a typical constrained target tracking problem. [0003] However, when dealing with the problem of constrained target tracking, conventional target tracking methods have obvious defects: first, the shape of the constraint trajectory is complex, and the traditional motion model mismatch is serious; second, the conventiona...

AI Technical Summary

Problems solved by technology

[0003] However, when dealing with the problem of constrained target tracking, conventional target tracking methods have obvious defects: 1. The shape of the constrained trajectory is complex, and the mismatch of the traditional motion model is serious; 2. The conventional method does not consider the prior information of the constraint conditions, resulting in information waste; 3. The tracking results are difficult to meet the constraints, and there is a performance loss

Such methods still fail to resolve the mismatch between conventional models and complex constraints

[0009] The above-mentioned methods are all based on the conventional unconstrained model to study the constraint tracking problem, and do not systematically solve the problem of model mismatch, and most of them are based on the assumption that the target has no maneuvering.

Images