ZED-F9R-Integration manual
Unlike automotive dynamic model, position, velocity, and attitude are not frozen when the e-
scooter is static.
The e-scooter dynamic model is not optimized for parking garages or longer GNSS outages.
Automatic wheel tick polarity detection should not be used. This is achieved by the
configuration item CFG-SFODO-DIS_AUTODIRPINPOL = 1
When the IMU is installed at the handlebar, the navigation performance is degraded for two
reasons: Firstly, the sensor fusion navigation filter assumes that the IMU shall be mounted
to a fixed orientation with respect to the vehicle compartment where the wheel speed is
measured. This assumption is broken when the IMU is installed to the handlebar while
getting the wheel speed data from the rear wheel. Secondly, the IMU measurements tend to
be noisier in the handlebar.
3.2.2.3 RLM dynamic model
Supported from firmware version HPS 1.21 onwards
The RLM dynamic model is used for applications with dynamics equivalent to those of a robotic
lawn mower (RLM). A robotic lawn mower is a low-speed vehicle with two pulling non-steering wheels
attached to the same axis and capable of turning at different rates. The RLM dynamic platform
model can be applied to any low-speed vehicle (max 3 m/s) with these characteristics. The RLM
dynamic model shall be selected by the configuration item CFG-NAVSPG-DYNMODEL = 11.
It combines the GNSS signal with the data from gyroscope, accelerometer and wheel ticks (or speed
measurements), more information can be found in the Solution type section. The RLM dynamic
model does not support degraded HPS mode, and requires wheel ticks or speed data to maintain
fusion mode.
To reach decimeter-level accuracy, the ZED-F9R requires RTCM stream at all times.
The RLM dynamic model only supports the User-defined IMU-mount alignment.
The receiver starts calibrating the sensors, once the following conditions are met.
• Speed is over 0.15 m/s.
• Heading accuracy is under 30 deg.
• Heading rate is less than 5 deg./s.
• RTK mode is fixed.
The INS filter is also allowed to initialize when the lawn mower is moving backwards. This is possible
since the customer is required to provide odometer data with polarity.
The RLM dynamic model can work with all the odometer inputs mentioned under the
Odometer interfaces section. However, it is recommended to provide wheel ticks with
polarity from both rear wheels. While enabling the configuration item CFG-SFODO-
COMBINE_TICKS = 1 .
It is also strongly recommended to configure the coarse wheel tick scaling factor by the
configuration item CFG-SFODO-FACTOR.
Automatic wheel tick polarity detection should not be used for all the odometer
inputs (wheel ticks / speed). This is achieved by the configuration item CFG-SFODO-
DIS_AUTODIRPINPOL = 1.
Unlike automotive dynamic model, position, velocity, and attitude are not frozen when the
lawn mower is static.
UBX-20039643 - R06
3 Receiver functionality Page 22 of 119
C1-Public
To Next Page
Loading...
Need help?
General
