DocumentCode
3234865
Title
Time optimal control of ultrasonic transducers for improved multiple object recognition
Author
Steiner, Gerald ; Watzenig, Daniel ; Schweighofer, Bernhard
Author_Institution
Inst. of Electr. Meas. & Meas. Signal Process., Graz, Austria
fYear
2004
fDate
24-25 May 2004
Firstpage
69
Lastpage
73
Abstract
Ultrasonic distance sensors are widely used in robotics and automotive applications because of their low cost, robustness and small size. One disadvantage of low frequency piezoelectric transducers operating in air is their small bandwidth. This results in poor spatial resolution when discriminating multiple objects within the directivity of the sensors. One method proposed in the literature is using amplitude modulation. Drawbacks of this method are very long pulse duration and the resulting high effort for matched filtering of the received signals. In order to detect multiple objects, a novel approach is proposed. It is based on time optimal control of the transmitting transducer. The optimal control sequence is determined by quadratic programming. The resulting pulse signal has a considerably shorter decay time compared to a conventional burst excitation. This allows a better resolution for the recognition of objects that are close to each other.
Keywords
amplitude modulation; object detection; object recognition; piezoelectric transducers; quadratic programming; time optimal control; ultrasonic measurement; ultrasonic transducers; amplitude modulation; automotive applications; low frequency piezoelectric transducers; multiple object detection; multiple object recognition; quadratic programming; robotics; robustness; spatial resolution; time optimal control sequence; transmitting transducer; ultrasonic distance sensors; ultrasonic measurement system; ultrasonic transducers; Automotive applications; Bandwidth; Costs; Frequency; Object recognition; Optimal control; Piezoelectric transducers; Robot sensing systems; Robustness; Ultrasonic transducers;
fLanguage
English
Publisher
ieee
Conference_Titel
Robot Sensing, 2004. ROSE 2004. International Workshop on
Print_ISBN
0-7803-8296-X
Type
conf
DOI
10.1109/ROSE.2004.1317617
Filename
1317617
Link To Document