UWB positioning method based on particle swarm optimization support vector machine and autonomous integrity monitoring

The invention provides a UWB positioning method based on a particle swarm optimization support vector machine and autonomous integrity monitoring, and the method comprises the following steps: 1, collecting UWB signal features in an LOS/NLOS environment as a training sample, training the training sample through employing a PSO-SVM, and obtaining an optimal classification model; 2, classifying the acquired dynamic positioning data by using an optimal classification model; step 3, carrying out autonomous integrity monitoring on the ranging values of which the category is LOS, and finding and rejecting abnormal ranging values; and a fourth step of resolving the optimized UWB original data by adopting standard Kalman filtering to obtain a resolving track of dynamic positioning. According to the method, UWB signal features and ranging values are used as data, an error recognition algorithm based on channel statistics and distance estimation is adopted, NLOS errors are recognized from the angle of signal features t