The RT3000 family measure position in the same way as most advanced Inertial and GPS Navigation Systems. The position is computed using a WGS-84 model of the earth and the Latitude, Longitude and Altitude are used for the calculations. This is the same co-ordinate system that GPS uses. An undulation table can be used to improve the Altitude accuracy. The altitude accuracy of the WGS-84 model is about 50m, using an undulation table improves the accuracy, compared to sea level. This feature can be disabled if required.
The outputs can be the Northing or Easting compared to a local grid, or an output of the X and Y position offsets for a local grid that is rotated. This makes it much easier to plot the position of a vehicle as it is driven round a test track. The position is measured to an accuracy of 2 cm or better for the highest accuracy product/mode.
The RT3000 family output the distance travelled over the CAN bus and using TTL pulses similar to wheel speed. Distance accuracy is typically 0.05% of distance travelled and has a drift of about 3cm/s.
The RT3000 family measure velocity in a North, East and Down co-ordinate frame. This can be rotated to the Forward and Lateral directions. The 2D or 3D speed of the vehicle can also be measured. The velocity can be measured to an accuracy of 0.05km/h or better in the highest accuracy product/mode. Because they contain a commercial GPS receiver, the velocity is limited due to export control restrictions to about 1800km/h (500m/s).
The RT3000 family measure acceleration using three servo accelerometers. These accelerometers are extremely accurate compared to non-servo accelerometers. The bias, scale factor and linearity are all exceptional. The Acceleration is measured in the body-frame, that is in the XYZ directions in which it is currently pointing in. It will rotate its own measurements to the body-frame of the vehicle so that the vehicle’s XYZ accelerations are output. It can also output the accelerations in a Forward, Lateral and Downward direction that is independent of the orientation of the vehicle. The Forward and Lateral accelerations are perpendicular to gravity, so they do not include gravity; they are useful for measuring lateral acceleration of vehicles since the are unaffected by gravity coupling with the roll of the vehicle. Acceleration is measured to an accuracy of about 1mm/s² in its most accurate mode and over a 1g range. The measurement range can be 10G or, on request, 30G.
The RT3000 family measure heading, pitch and roll angles. These angles define the yaw from North (heading) the incline of the front of the vehicle compared to the back (pitch) and the sideways tilt of the vehicle (roll). The angles output are the Euler angles. Three consecutive rotations are required to get from the North, East, Down directions to the XYZ directions of the vehicle. The first rotation is by the Heading (about the Z-axis), then the second rotation is by the Pitch (about the new Y-axis) and finally by the Roll (about the newest X-axis). Euler Angles are not used internally since these can be unstable at certain orientations (when pitch is 90 or –90°); instead quaternions are used. The Heading can be measured to 0.1° (0.05° using a wider 4m separation of the dual antennas) and Pitch and Roll to an accuracy of about 0.03° in its most accurate product/mode.
The RT3000 family measure angular rate using three silicon MEMS angular rate sensors. These angular rate sensors are very accurate for MEMS devices. The Angular Rates are measured in the body-frame, that is in the XYZ directions that it is currently pointing in. It will rotate its own measurements in to the body-frame of the vehicle so that the vehicle’s XYZ angular rates are given. It will also output the angular rates in a Forward, Lateral and Downward direction that is independent of the orientation of the vehicle. The Lateral angular rotation is the Pitch Rate; the Downward angular rotation is the Yaw Rate whereas the body-X direction is the Roll Rate. The measurement of Angular Rate is to an accuracy of about 0.005°/s. The measurement range is 100°/s or, on request, 300°/s.
Angular Acceleration tends to be a very noisy quantity because of vibration. The RT3000 family can measure Angular Acceleration and filter it. The Angular Acceleration outputs are available on the CAN bus. The Yaw, Pitch and Roll Angular Accelerations are computed, filtered and output.
Track Angle, Velocity Vector Angle, Course over Ground
The Track Angle is the direction that the vehicle is moving in, rather than the direction that it is pointing in. The Track Angle is derived from velocity. It is the same angle that GPS outputs and it bears no relationship to the direction that the vehicle is actually pointing in. The accuracy of the Track Angle depends on the speed. This is because the velocity accuracy is fixed, so the faster the vehicle moves, the more accurate the Track Angle becomes. Typically the Track Angle is accurate to about 0.05° at 50km/h in its most accurate mode.
The RT3000 family compute the Slip Angle by comparing the Heading Angle to the Track Angle. The difference between the two is the Slip Angle. It is a measure of how much the tyres are slipping sideways. The Slip Angle is very important for chassis engineers and tyre manufacturers. The RT3000 family give very high bandwidth slip angle compared to optical sensors and results compare very well with simulation systems up to and above 5Hz. (5Hz normally being the maximum frequency of interest for chassis work, rather than the limit of the products). The RT3000 can measure the Slip Angle to an accuracy of about 0.15° in its highest accuracy product/mode.
All GPS based products know time very accurately. The GPS second is the same as the UTC (or GMT) second that is maintained as a reference in Paris. There is an offset between GPS time and UTC time that needs to be taken in to account for some applications. Every RT3000 family product samples at exactly the same time (to within about 50ms) and the samples are synchronised to the GPS clock. This makes the output samples exactly 100Hz by definition and there is no drift or error in this timing. An output of 1 pulse-per-second can be provided that can be used to synchronise other devices to UTC or GPS time. A 100Hz sample outputcan be provided that is synchronous to the GPS clock and to the product's outputs.
The RT3000 family generate a whole host of Status Information that can be used to monitor the accuracy and performance of the outputs. These provide a quality control measure to ensure that the data is always as accurate as required by the application.