User Manual

Table Of Contents
Page 37 of 129 54A0028-6 042019 © LeddarTech Inc.
filter out the false measurements at the application level. For example, this can be useful in
applications that require long detection ranges or detection of small or low reflectivity targets.
Smoothing
The smoothing algorithm increases the precision of the measurement at the cost of the LeddarVu
module reactivity. The history length of the filter is defined as a function of the measurement noise
level. It also changes according to the oversampling and accumulation settings.
The history length of the averaging filter can also be adjusted by a parameter ranging from 16 to
16. Clear the Enable check box to disable smoothing. Higher values increase the module precision
but reduce the module reactivity. An example of the behavior of the measurement smoothing
algorithm is depicted in Figure 17.
Figure 17: Measurement smoothing example
The red line represents the true target distance; the blue curve corresponds to the target distance
measured by the module without smoothing, while the green curve is the smoothed measurements.
One could notice the measurement precision (standard deviation) is dramatically improved by the
smoothing algorithm.
NOTE: The smoothing algorithm is recommended for applications that need highly precise
measurements of slowly moving objects. For application that tracks quickly moving objects, it is
advised to decrease the value of the smoothing parameter or to disable the smoothing algorithm.
0 20 40 60 80 100
14.98
14.99
15
15.01
15.02
15.03
15.04
15.05
Sample
Distance (m)
Estimated
Measured
True