Results: 1-Dimensional Location. Results: 1-Dimensional Location. Determination of Source Position. Results: 1-Dimensional Location

Transcription

1 Mobile robot using different senses Motivation Senses for Robots ISOEN ¾ Sight (Cameras) ¾ Hearing (Microphones) ¾ Touch (Tactile sensors) ¾ Smell (Gas sensor system) ¾ Taste (Electronic tongue) M. Wandel*, A. Lilienthal+, A. Zell+, U. Weimar* *ipc, AG Weimar +WSI-RA, AG Zell www-ra.informatik.uni-tuebingen.de Extending the possibilities ¾ Electronic Watchman can oversee chemical stocks and detect leakages Overview Robots with Gas Sensors: Applications Biomimetic Path Finder Virtual Umbilical Repellent Marker Motivation ¾ State of the art Experimental set-up ¾ Mobile robot ¾ Sensor system Potential Applications Trail following ¾ Trail defined life time Marking ¾ Mark of Cleaned Floor Source Detection Experiments ¾ Direction finding ¾ -Dimensional experiments in a corridor ¾ -Dimensional experiments in large room Rotating Robot Short Distance (Duckett et al.) Three dimensional Odour Compass Findings Conclusions Outlook Defined air stream (Nakamoto et al.) ¾ Detection of odour plumes Plume tracking (Russel et al.) Defined air stream needed

2 Robots with Gas Sensors: Applications Experimental: Sensor System Source Detection Potential Applications Biomimetic Direction finding Trail following Path Finder ¾ Rotating Robot ¾ Trail defined life time Short Distance (Duckett et al.) Virtual Umbilical Compass Marking ¾ Three dimensional Odour Repellent Marker ¾ Mark of Cleaned Floor VOCmeter Vario Commercially available Lightweight, small Low power consumption (V DC supply possible) Defined air stream (Nakamoto et al.) Source Detection Detection of odour plumes RS- interface ¾ Direction finding ¾ Plume tracking (Russel et al.) Rotating Robot Short Distance (Duckett et al.) profile Three dimensional Odour Compass Defined air stream needed Convection / Diffusion ¾ Map Building Various sensors types available (MOX, QMB, U/I) Defined air stream (Nakamoto et al.) Up to sensors, connected with thin flexible cables No Defined Airstream needed ¾ Detection of odour plumes Plume tracking (Russel et al.) Defined air stream needed 5 7 Experimental: Mobile Robot Arthur c m -Dimensional Location Mobile Robot ATRV-Jr Skid Steering Standard PC inside Ultrasonic sensors Wireless Ethernet Locations Additional Sensors Laser range finder Stereoscopic camera System Gas sensor system mounted inside the robot, behind front window ¾ No ventilation! ¾ No personnel traffic Experiments Analyte used: Ethanol Recordings ¾ odometry, sensor system, laser range finder, ¾ temperature, humidity (offline)

3 Results: -Dimensional Location Results: -Dimensional Location n [ppm] - n [ppm] Results: -Dimensional Location Determination of Source Position n [ppm] - c [ppm] - c [ppm]

4 -Dimensional Location Comparison Summer - Winter Unventilated One side of the room is a window front The automatic source was placed in the middle of the room The robot s path describes a rectangular helix Winter signal [a.u.] signal [a.u.] Summer y 5 5 x 5 Results -Dimensional Location Peaks / Position, signal [a.u.],5 Y [m],,5,,5 Summer Winter, 5 X [m] 7

5 Time of Occurrence Conclusion signal / a.u. Signal (baseline) Signal (peak) Online computed baseline Offline fit t open, predicted topen, time / s real In a -dimensional environment the position of the source can be estimated within a range of m In - and -dimensional environment the time of occurrence of the leak can be computed with good precision The measured concentration profile requires time consuming search strategies e.g. map building, this is possible because of the stable concentration profile 7 9 Findings Outlook Signal nearly independent of stick position Simple mounting of the sensor, without fan, without pumping unit, sufficient Meaningful measurements only during movement Even small leaks can be detected Implementation of a feedback from the sensor signal evaluation to the driving programme Larger rooms Classification of the analyte Using a sensor system with different sensors including QMBs and pattern recognition software Testing in real world environment

6 Acknowledgements Landesforschungsschwerpunktsprojekt Baden-Württemberg Sensing organs for mobile robots Project partners: Wilhelm-Schickard-Institut für Informatik, AG Zell (coordinator) Zoologisches Institut, AG Schnitzler Institut für Physikalische und Theoretische Chemie, AG Weimar Thank You