Expo GPS Documento

8/14/2019 Expo GPS Documento

1/16

Christian Leonardo DaJonathan Jindo Ni

Telecommunicat


2/16

The main objective of the work presented in this pa

develop and evaluate a method for measuring averuneven forest terrain with an 8WD forestry machydrostatic-mechanical transmission. Additionally, to measure and analyse slip of individual wheels is In a previous study, the forest machines

transmission measurement was used and combineddata for measuring mobility parameters. We use a senhanced, technological approach specially adaptemeasurements.


3/16

The American GPS (Global

Positioning System).

GLONASS

Navigation Satellitfrom Russia.


4/16

Fig. 1.Example of a typical forwarder, a machine that carriescut timber from the forest to a landing.

One of the most commonly usedforestry isthe forwarder.

The carrying ca

normally between tonnes and commonforestry the macheight driven wheels improved comfort aon soft ground.


5/16

Olsen and Wsterlund reported studies on an exforwarder supplied with torque-measuring axles torolling resistance and torque distribution for an 8Wmachine. After a short and slow travel straight forlocked differentials, the front wheels started to sl

rear wheels were pushing the machine forward (Fig.


6/16

Fig. 3 Torque measurements (kN m) on each drive shaft to thewheels when driving straight ahead during 20 s on gravel groundwith locked differentials.


7/16

Shoop et al. measured traction and slip on a 4WD

driving in soft terrain and snow. Wheel slip was measurea free rolling axle or a single 5th test wheel on the SAtesting car. A fifth wheel as a sulky was successfulMohr & Eriksson when studying an 8WD forwarder driviground.

Angelova et al. suggested a framework for learning to for a Mars rover. The terrain in front of the vehicle wausing stereo camera images. Visual features for terraiwere used to learn a nonlinear model of the corresponslip.


8/16

efinitionsThe actual vehicle velocity is denoted by V.

The longitudinal slip of a wheel is the difference between wheel speand V .

V may for example be computed from consecutive estimated positioa DGPS, as is done in this paper.

slip coefficient


9/16

evelopment of a slip measurement methodologFor this study, experiments on four different ground types were perforasphalt, hard gravel, sand, and forest terrain. On level areas of dry asand hard gravel, the forwarder was accelerated from stands till to constant speed.


10/16

Speed measuring technique and signal filteringThe speed of the machine was measured via its standard odometer, wh

used a magnet and pulse encoder on the outgoing shaft of the transmihydraulic motor.

Hence, the wheel speed vw see as an average speed for all eight wheelThe wheel speed is sampled in 2.5 Hz. The vehicle velocity V was estimfrom consecutive position data from a real-time kinematics differential(RTK-DGPS) with a sampling rate of 5 Hz, and is henceforth denoted vg

To reduce errors due to uneven terrain, we used all three dimensions inGPS data to calculate the velocity.

For calculation of individual wheel slip, the corresponding transformatiwould have to be done for each wheel. The timestamps for the sensorreadings differed between the two sensors, so the values for vgps and vwere linearly interpolated to have matching timestamps.


11/16

Difference between wheel and GPS speed (vw and vgps

respectively) (upper pane) and the actual values for vwandvgps (lower pane) when driving the track Asphalt 2.


12/16

Observed slip

Mean (l) and standard deviation (r) for slip values (%) when driving straforward at constant speed on asphalt and gravel. N is the number ofobservations.


13/16

Position data delivered by the RTK-DGPS system for a nonmoving receivThe noise (standard deviation 1.98 mm in the shown example) leads to non-zero speed estimate based on consecutive positions.


14/16

iscussion and conclusions

The field study verifies that the slip of an 8WD machine can be estimatcomparing the GPS-based speed with the wheel speed, both on even arough terrain.The method can be a useful tool for the improvement of the transmisschain of a 6WD and 8WD machine, decreased fuel consumption andimproved trafficability of all kind of off-road vehicles.


15/16

In general, computation of slip at low speed is numerically sensitive terrors in both vgps and vw. This is illustrated in Fig. 10 which shows the calculated slip analytically (by applying Eq. (1)) depends on DV atvarying wheel speeds when the true slip is zero.


16/16

Advanced Communications

Expo GPS Documento

Documents

Transcript of Expo GPS Documento