Multiple Maneuvering Targets Tracking Algorithm And Simulation Platform Developing

Multiple maneuvering targets tracking problem is a research focus of the field of target tracking at present, and it has extensive application perspective in military and civil industries, such as air warning, battlefield surveillance, air traffic control, and navigation systems, etc. With the rapid development of science and technology, modern track environment has undergone significant changes. Multi-targets tracking and data association technology faced strong challenges. This paper is concerned with two aspects of multiple maneuvering targets tracking problem through the "11th Five-Year Plan" weapons and equipment pre-research projects, those are maneuvering target tracking and data association. The main work and achievements are as follows:First, an overview of the basic principles of maneuvering target tracking is presented. The basic elements of maneuvering target tracking problem are summed up, including the target motion models and the commonly used filters for maneuvering target tracking algorithms. The paper also analyzed the characteristics of the various filtering algorithms.Second, to implement the problem of multiple non-maneuvering targets data association and track maintenance in cluttered environment, the traditional associating methods such as joint probability data association (JPDA) and new associating methods such as graph-based and biology-based association methods are studied. And two kinds of data association algorithms including the JPDA and m-best MHT are implemented. Taking into account the high computational complexity of the m-best MHT algorithm, two improved m-best MHT algorithms are proposed by reducing the dimension of the row vectors and column vectors of the cluster matrix. Respectively, the proposed algorithms perform better in the environment which has large number of targets and measurements. Monte Carlo simulation results show that the algorithm can shorten the time with maintaining the filtering accuracy and association accuracy.Third, to implement the problem of multiple maneuvering targets data association and track maintenance in cluttered environment, two kinds of algorithms including interactive multiple model joint probabilistic data association (IMMJPDA) and interactive multiple model multiple hypothesis (IMMMHT) have been researched and implemented. And a improved IMMMHT algorithm are proposed by grouping the measurements, and greatly eased the problem of enormous calculation caused by splitting the high-dimensional cluster matrix. By two and three maneuvering target tracking simulation results show the effectiveness of the above algorithm.Finally, a simulation software platform is designed and developed through data flow and components in Visual C++ 6.0 development environment. In order to simulate, test and evaluate the multi-sensor multi-target tracking algorithm. The structure and the function of the platform are introduced, and the problems of software design and data structure design are discussed. The effectiveness and practicality of the platform is tested by a instance running in the platform.