USING AN LB301/LB201 LONG RANGE LASER DISPLACEMENT SENSOR FOR SHAPE-SENSING OF A BLACK RUBBER MOULDING BEAD BAG

Dominic P.K. Cheng Rehabilitation Engineering Centre The Hong Kong Polytechnic University

ABSTRACT

The LB301/LB201 laser displacement sensor produces spurious displacement readings when it is used to scan a black rubber moulding bead bag. If each scan is post-processed with an appropriate low-pass digital filter, the resulting displacement values agree very well with that obtained using a mechanical tracer. The laser sensor can measure a moving moulding bag driven at a horizontal velocity of 20mm/sec with acceptable accuracy. At this velocity, it takes under 10 minutes to trace a 460mm x 460mm bead bag.

BACKGROUND

The LB301/LB201 laser displacement sensor has a range of ±100mm at a reference distance of 300mm. At this reference distance it has a light spot size of 1.2mm x 25mm. A previous study found that the sensor can measure contours of moving spherical objects ranging from a black rubber basket ball to a small blue rubber ball (28.4mm dia.) with acceptable accuracy for special seating. However, because the sensor measures displacement by sensing the angle between an incident and reflected beam, it was found that acceptable accuracy was obtained only within a subtended angle of about ±70 from the apex for spherical objects. (1)

OBJECTIVES

To measure the performance of the laser sensor in shape sensing of a black rubber moulding bead bag impressed with the contour of a thin adult.

METHOD

Determining safe working horizontal velocity (Fig. 1) The LB301 sensor was mounted on the Z-axis (vertical-axis) of a vertical machine center (VMC). The visible emitted beam was aligned vertical to the table of the machine. The bead bag, with the impression of a ball (28.4mm dia.), was placed flat on the table which was driven to move under the sensor in the X-direction (horizontal) at pre-determined velocities. For each velocity, the sampling rate for the laser sensor output was adjusted to capture a sample per 0.2mm of displacement along the X-axis.

Verification of displacement measurements (Fig. 2) The bead bag was impressed with the contour of the buttock of a thin adult. The bag was driven along the X-direction at a velocity of 20mm/sec. A scans was taken, across the ischio-tuberosites. Vertical displacement measurements were taken at intervals of 0.2mm of displacement along the X-axis. Vertical displacements were also sampled every 5mm of horizontal displacement using a touch probe, also mounted on the Z-axis. The probe had a 4 mm dia. steel ball at the tip

RESULTS

Safe working horizontal velocity (Fig. 3) Essentially the same displacement values were obtained from a velocity of 120mm/min through 4,000mm/min. The displacement values for 6,000mm/min and 12,000mm/min were clearly off. To be cautious, a velocity of 1,200mm/min (i.e. 20mm/sec) was chosen for the remaining measurements in this project.

Verification of displacement measurements (Fig. 4-5) Fig. 4 showed that, except for the rapid spurious displacements, the vertical displacements measured by the laser sensor corresponded well with that by the touch-probe. The spurious displacements varied too rapidly to be real from an anatomical view point. These were clearly an artefact from the sensor in response to the lumpy texture of the moulding bag. The beads inside the bag produced small but severe contours on the surface. As stated earlier, good accuracy was obtained from spherical objects within a subtended angle of about ±70 from the apex. If the output of the laser sensor was taken to be a low frequency signal contaminated with high frequency noise, the noise could be filtered out using a low-pass electrical filter. Fig. 4 showed that after processing by a digital 2-pole low-pass Butterworth filter, the resulting signal corresponded very well with that from the touch-probe.

DISCUSSION

From the results obtained, the laser sensor is suitable for shape-sensing of a black rubber moulding bead bag for special seating for two reasons. . At 20mm/sec, it requires only about 10 minutes to scan a 460mm x 460mm (approx. 18" x 18") bead bag if only cross-sections 20mm apart are traced. The displacement accuracy is quite acceptable if post-processing of the displacement values with a digital filter is performed.

Since the performance of the sensor is done empirically, it is prudent to use it with caution and check its results with known means when in doubt.

REFERENCE

1. Cheng, Dominic (1995) Evaluation of a laser displacement sensor for use in special seating. RESNA'95, 267-9.

ACKNOWLEDGEMENT

This work was supported by The Hong Kong Polytechnic University.

Dominc P.K. Cheng REC, The Hong Kong Polytechnic University Hung Hom, Kln. Hong Kong email rcdpkc@polyu.edu.hk