THEME 3: MULTI-SENSOR INTEGRATION

GPS/Pseudolite/INS Integration

Figure 3: DD residuals in four satellites pairs in the GPS/INS case

Figure 4: DD residuals in four satellites pairs in the GPS/PL/INS case

Results of the experiment were published in:

LEE, H.K., WANG, J., RIZOS, C., & GREJNER-BRZEZINSKA, D., 2004. Analysing the impact of integrating pseudolite observations into a GPS/INS system. Journal of Surveying Engineering, 130(2), 95-103. (Download PDF)

In addition to the research on this new positioning and navigation system design, GPS carrier phase processing techniques (ambiguity resolution and cycle slip detection & identification) have been developed for maintaining high accuracy of the GPS/pseudolite/INS system. The GPS carrier phase measurements with correctly estimated integer ambiguities must be utilised to update the system integration filterÕs states. The occurrence of a cycle slip that is undetected is, however, one major error source that can significantly degrade the filterÕs performance. Therefore this research has developed not only an effective procedure to increase the reliability and speed of ambiguity resolution by means of additional pseudolite and INS measurements (see Figure 5), but also an algorithm which can effectively detect, using Cumulative-Sum (CUSUM) test, and identify the cycle slips.

Figure 5: The ambiguity resolution procedure

A series of simulation studies based on Ambiguity Resolution of Dilution of Precision (ADOP) values were carried out to investigate the performance of the proposed ambiguity resolution procedure (see Figures 6 & 7).

Figure 6: Impact of pseudolites on ADOP (as a function of the number of pseudolites)

Figure 7: Impact of INS-predicted position error on ADOP during signal blockages (GPS/PL/INS systems)

On 23rd April 2003, land vehicle experiments were carried out to evaluate the ambiguity resolution (AR) and cycle slip detection & identification performance for the integrated GPS/pseudolite/INS system.

Figure 8: Experiment set-up

Figure 9 shows the time-to-fix of L1 carrier phase ambiguities after complete GPS/pseudolite signal blockage. The results manifest two important characteristics. First, the GPS-only solutions are the poorest in all cases. Second, the AR performance of GPS/INS and GPS/pseudolite/INS are similar after a short signal outages (up to 10 seconds). However, the AR performance is significantly improved by employing pseudolite measurements for outages of 20 seconds or more. Special attention should be paid to Case I, because this is where the most significant improvement was obtained. Hence it is demonstrated from the results that the proposed AR procedure based on GPS/pseudolite/INS integration makes it possible to resolve the ambiguities within a couple of seconds if the outage is relatively short. Moreover, the AR performance can be considerably enhanced even in the case of a blockage of 50 seconds.

Figure 9: Time-to-fix L1 carrier phase ambiguity after different lengths of signal blockage

Figure 10 illustrates the cycle slip decision values and two-sided CUSUM values for pseudolite measurements. The top graphs in these figures depict the decision values with the introduced cycle slips in the raw measurements. The second and third graphs show that the CUSUM values remain at zero when no cycle slip occurs. At the occurrence of a cycle slip, the detection threshold is exceeded and the cycle slip determination process identifies the affected double-differenced observations. It is seen from the results how well the algorithm performs in the data processing as slipped carrier phase measurements are successfully detected and identified without missing any cycle slips, despite some slips occurring on two successive epochs.

Figure 10: CUSUM test for pseudolite measurements to detect & identify the cycle slips

More details on this research issue can be found in: