Background

It is general awareness that a dual antenna system can offer better accuracy and test result than a single antenna system. This article aims to give a quantitative analysis to demonstrate how much difference will there be between dual antenna system and single antenna system, under the same testing condition.

Test set-up

• Equipment

For this test, we used a RT3000 v.3 with dual antenna set-up. Since data collecting was carried out on open road, NTRIP was enabled for the unit to reach RTK level accuracy.

The RT and both antennas are properly mounted onto the vehicle.

To control all variables in this comparison test, we will use the same data set collected from this run, and post process it in two configurations:

1. Dual antennas configuration

2. Single antenna configuration

This way, the only variable between the two configurations is the number of antennas used. 

• Data collecting route

Here is a bird eye view of the data collecting route:

The duration is about 1 hour and 20 minutes.

The route contains a good warm-up at the beginning, and has a good combination of different GNSS conditions.

Data analysis

Accuracy

In accuracy comparison section, we synchronize the accuracy data from both configuration and use function:

Compare_value = Config2_accuracy - Config1_accuracy;

Since Config2 is the single antenna system and Config1 is the dual antenna system

Compare_value > 0 means Config1(Dual antennas) has better accuracy

Compare_value < 0 means Config2(Single antenna) has better accuracy

All the plots below are showing Compare_value for different accuracy.

• Heading accuracy comparison

From the plots we can see that we are getting positive value from the function.

Compare_value > 0 means Config1(Dual antennas) has better accuracy

This indicates that Dual antenna system has better heading accuracy, and in the histogram we can see the heading accuracy increases substantially, mostly more than 30 percent in dual antenna system.

• Position accuracy comparison

As the plot shows, dual antenna system has better position accuracy, and has around 10% better position accuracy than single antenna system.

• Velocity accuracy comparison

As the plot shows, dual antenna system has better velocity accuracy, and has around 10% better velocity accuracy than single antenna system.

• Slip angle accuracy comparison

As slip angle accuracy is a function of heading, it is expected that dual antenna system provides better slip angle accuracy. From the plot we can see around 30% of improvement in slip angle accuracy from dual antenna system.

Heading accuracy vs Vehicle velocity

Low vehicle velocity tends to be trouble for heading accuracy for INS. Here we plot heading accuracy against vehicle forward velocity. To make the plot more readable, we divide velocity value by 100 to scale the data. As the original units for velocity is m/s, now in the plot it is shown as cm/s.

It's easy to see between 500s and 100s the vehicle is at very low speed(or stationay), and single antenna system loses heading accuracy very quickly, while dual antenna system is able to maintain a consistent good performance.

Impact on Acceleration

Here we compare the 20 points moving standard deviation of forward acceleration from both system. System with lower moving standard deviation is considered to deliver a more consistent result with less noise.

 Again, we use function: 

Compare_value = Config2_Standard_Deviation - Config1_Standard_Deviation;

In this case:

Compare_value > 0 means Config1(Dual antennas) has better acceleration data

Compare_value < 0 means Config2(Single antenna) has better acceleration data

Here we plot the Compare_value from the function, and we can see that the value looks evenly distributed above and below zero.

To get a better view of the situation, we count the number of Compare_value greater than 0, and count the number of Compare_value smaller than 0; then put them in a pie chart.

This confirms that there is very little difference in the moving standard deviation from both systems, which means in this test, dual antenna system have similar quality of acceleration data as single antenna system.

Conclusion - the benefit of Dual antenna system

• Enable the capability of static initialization 
• Better Heading accuracy (more than 30% improvement from single antenna system)
• Better position accuracy (more than 10% improvement from single antenna system)
• Better velocity accuracy (more than 10% improvement from single antenna system)
• Better consistency in heading accuracy at low speed

Note: The percentage of improvement only apply to this specific test. This value could vary in different testing condition.