Application of Mean Field Annealing Algorithms to GPS-Based Attitude Determination

This paper researches the attitude determination of a body based on GPS carrier phase measurement, and the algorithms used is the mean field annealing neural network(MFANN), which has the advantages of high accuracy and rapidity. The MFANN approach is a combination of the competitive Hopfield neural network and the stochastic simulated annealing technique, which is efficacious in resolving the optimal attitude determination. Firstly, the fundamental principle of GPS attitude determination is described, then a test platform of attitude determination is set up, and the MFANN algorithm is verified to resolve integer ambiguity and azimuth angles. Lastly, the experimental example is presented by means of MFANN, and it shows that this method is effective.