The 28th Annual Conference of the Japanese Society for Artificial Intelligence, 2014
- 1 -
J4ðÀūŠőŔĦijĶłŬŠŮŕí9ÍÑ
Onboard Sound Recognition using Small Aerial Robot
> Òº
6
Kensho Miyoshi Koichi Hori
}=@=@çK@«£¿¤ABK@Ik
Department of Aeronautics and Astronautics, The University of Tokyo
Recent and rapid advancement of robotic technology is bringing aerial robots closer to tasks and applications. Natural interaction of aerial robots is expected to be applied in wide areas from disaster countermeasures to entertainment. However there was only a method of using external microphone to realize audio-based interaction between human and aerial robots. We propose audio-based interaction using onboard microphone to realize closer and directer interaction.
1.
ð À ū Š ő Ŕ ģ â ģ ħ [ ĥ Ľ Ŭ ŏ Ũ ń ʼn ŧ Ŭ ůNatural Interaction, NIŰĦãęĶ£ĒÖOPĔÀĸķĶijĎĦĥĵ, FH¨đĴŀŬŏŮŒĽŬţŬŔĮĢNPčʼnŮŬĢħXĒ
wTėķġčĶ.ó V|£ĦĨðÀūŠőŔǤđĴbďĶq
ĹčĝıħĒ<ĔįĴķĶĒ, z£ĨðÀūŠőŔħĊÅ
ÆċĢĨĥĔĊ¼ÆċĦãęĶ£ĢČĶ.
ŦŮňĒhM4šĽńĦí9ĹĘ, í9ÍÑĦijĵðÀū
ŠőŔĹ³j´ęͧģčğĝí9ĽŬŏŮśĿĽŋĦĠčġ
ĨV|Ħ£Ēĥėķġēĝ. ĘđĘ, ðÀūŠőŔ¾Ēí9
ōŬŇĹďĶĖģĦijĶ-/8ģħĽŬŏŨńʼnŧŬħ)½Y
ĦĠčġĨĮĞĮĞxá`ĢČĵ, z£ĨĖħ)½YĹd
ęĶıħĢČĶ.
©å, ðÀūŠőŔģħNIĦĠčġãÙ£Ĺŭ¸
GĘ, łŬŠŮŕí9ōŬŇĦijĶí9ÍÑ, $ĪĜķĹčĝ
ĽŬŏŨńʼnŧŬĦĠčġDÂĘĝŜūŔŏĽŜĹ ďġÐpęĶ.
2.
ðÀūŠőŔħNI ĦãęĶV|£ĦĨ, ŊĿŋŐťŮĥĤ
ħŝŮŋħĻŜūŮŐģ, 9ĥĤħí9ŝŮŋħĻŜūŮ
ŐģčĎģĥĶųĠħ^ĒČĶ. sĦ, ĜķĴħ$Īí
9ōŬŇÜĒłŬŠŮŕĢČĶđ,đģčĎųØĵĦïĒĥė ķĶ. HTA(Heavier-than-air)4ðÀūŠőŔĦãĘġ, ĜħV|
£ĹęĶ.
MonajjemiĴĨńĻŕņŜŏŮģĜħłŬŠŮŕŃţŨĹčġ
ŊĿŋŐťŮĦijĶĽŬŏŨńʼnŧŬĹDĘĝ[Monajjemi 2013].
Ng ĴĨńĻŕņŜŏŮģ˵4ŃţŨĹčġŊĿŋŐťŮĦij
ĶĽŬŏŨńʼnŧŬĹDĘĝ[Ng 2011]. ¹ħ^Ĩ, ˵4
ŃţŨĦĭðÀūŠőŔħŞĽūŮŕ§ħŋŞőńĦ¬Ĺ(
ĕĶo, ðÀūŠőŔĨĜħµĦãĸĴĚ-/8ģ
ŝŮŋħĽŬŏŨńʼnŧŬĒ)½ĢČĶ. U¹ħ^ĢĨ,
'W$ĪĒðÀūŠőŔģìėķĝʼnŋŒŢĢÀĸ
ķĶĝİ, ðÀūŠőŔħŋŞőńĦ¬Ĺ(ĕĶĖģĒĥč. Ę
đĘĥĒĴ, ŦŮňĨ˵ėķĝŃţŨħĢħįðÀūŠőŔ
ĦHĘĻńʼnŧŬĹÀĎĖģĒÌėķġĐĵ, âĦģğġħĽŬŏ
ŨńʼnŧŬħ)½ĥî5ĒŃţŨħÅÞĦæCėķĶ.
QuigleyĴĨ1C·ħJ4ðÀūŠőŔĦHĘ, PDAůhMZ
7¦yŰĹčĝí9ĦijĶÚéjĹÀğĝ[Quigley
2004]. PDAĨ3Ģėķ, PDAĦijğġWĴķĝí9Ĩ
ņšŬŕĦ:fėķ, ņšŬŕĨðÀūŠőŔĬ³×ėķĶ.
Ėħ^ĨŦŮňĒðÀūŠőŔĹÚéj´ęĶĢv!ĢČ
ĶĒ, oĢðÀūŠőŔ¾ĦĨí9ōŬŇĨÂėķġč
ĥčĝİ, ðÀūŠőŔĨ¾Óħ-/8ħí9Ħ%XęĶĖ
ģĨĢēĥč.
ħijĎĦHTA4ðÀūŠőŔħNIĦĐčġ, ðÀūŠő
Ŕ¾Ēí9ōŬŇĹgÔĘ, Ĝħ-/8ģĽŬŏŨńʼnŧŬĹ
ÀģčğĝĻŜūŮŐĨ ĤįĴķĥč.
3.
#!$ "
]ĉĒDĘĝħĨ, ðÀūŠőŔħłŬŠŮ๥ńĹč
ĝí9ÍÑ$Ī, ĖķĹčĝ-/8ģħĽŬŏŨńʼnŧŬĢ
ČĶ. ˵4ČĶčĨhM4šĽń§ħœŘĽŋĹęĖģĒĥ
čĝİ, Cħj´¹ĹÛīĖģĒĥč. ŀŬŏŮŒĽŬţŬŔ§
ħÞĢĨ, ÃmĒ+rĦ#ĢēĶSħĽŬŏŨńʼnŧŬĒ
)½ĦĥĶ. Fl §ħʼnŮŬĦĐčġĨ, -/ħ
Z7Ħďí9Z7ħ&ëĦčĴķĶĖģǏďĴķĶ.
4.
z©ĢĨ, ]ĉħDÂĘĝŜūŔŏĽŜĹęĶ.ó ðÀūŠ
őŔģĘġ, {Ĵħ^ħĩĴŇĽŌńĻŕņŜŏŮ[Konomura
2013, 2014]Ĺčĝů01Ű. zĨŤŮŏđĴHÈħŤŮŏĮ
ĢĒ12cm, ÝßĨ70gĢ, ŎśŔľĿĻʼnŋŒŢĨLinuxħŜŨ
őŔśŁŮŢĦªėķġčĶ. šĽńĨœŊŏŪʼnũņŬšĽ
ńūśŁŬů02ŰĹčĝ.
Ù°Ŵ>Òºõó }=@=@çK@«£¿¤AB K@Ikó6£EõóāýĈăąüýÿûĂąüă÷øýĀøùõĆõćôĆăÿĈăõøúõþĄó
2L3-4
The 28th Annual Conference of the Japanese Society for Artificial Intelligence, 2014
- 2 -
4.1
ìèrĦĨ, ðÀūŠőŔ¾ħŗĽŌ;Ħí9Ĺä
FęĶıħĒĥč. ìèrħí9ÍÑÜĦĨ, łŮŜŬŎŮŋ
ħ=ÎRÙ±í9ÍÑŀŬŊŬJulius[Lee 2009]ĹcĘ
ĝ. JuliusĨũĻŪŏĽŢħí9ÍÑĹDÀęĶĖģĒĢē,
íÕtIJÊÎŤœŪ§Ĺ¾Ħ®įuďĶĖģĒĢēĶ.
z£ĦĐčġ, ŦŮňÕtĨņšŬŕ¶ĐijĪ, ðÀūŠő
Ŕħ. ħ"ÎĢ\ėķġĐĵ, 1). ħį, 2)ņšŬŕħį,
3). +ņšŬŕģčĎ¢ïħnĹÍÑęĶ. DêĦņšŬ
ŕĹDÀęĶħĨ 3)ħí9ĒWĴķĝģēħįģĘġčĶ.
ĖķĨ-/ħñíħÏħH¨ĢČĶ.
4.2
ðÀĦĨ, ńĻŕņŜŏŮ¾ħŜūŞŨĦijĶñíĦijĶ
ŗĽŌĒ?2ęĶ. ĖħñíĹ-mÉ~ęĶģ, 0 3 ħijĎĦ
âħ9ħ-mMĦÖč 100Hz1kHz ČĝĵĦ=ēĔL
ĘġčĶĖģĒĸđĶ. o, 3kHzÖħ-mMĦĨŗĽŌĒ
ČĮĵÄĴķĥč. ĜĖĢ, 3kHz¡QħíĒěĶşĽőŋŪĹ
čĝDòĹÀğĝ. í9đĴFIR śļŪŏů0 5ŰĹč
ġŘŬŕřŋśļŪŏĹđĕ, ¥ãmĦĨŃĽňŮ¥ãmĹč
ĝ. ¯, ŜūŞŨíĦijĶ=ēĥŗĽŌĒČĶ, şĽőŋŪíĹ
»ēĕĶĖģĦ\Ęĝ. şĽőŋŪíģèņšŬŕĹHXė
ěĶĖģĦijĵ, ðÀħðÀūŠőŔĦłŬŠŮ๥ńĹč
ĝèaĒ)½Ħĥğĝ.
5.
łŬŠŮ๥ń$ĪMCUĦijĶí9ÍÑĦijĶðÀūŠő
ŔģħĽŬŏŨńʼnŧŬĹDĘĝ. ŀŬŏŮŒĽŬţŬŔIJŖŚŅ
ŮʼnŧŬĥĤħÞĢ, ðÀūŠőŔģâħijĵÖĔeĥĽ
ŬŏŨńʼnŧŬĹ)½ĦĘġčĔĖģǏďĴķĶ. Įĝ, Z7&ë
ħÇđĴĨ, V|ĺĦ£ĒÀĸķġčͤiůZ
7ŰĦď, -mMħæĨČĶıħħí9Z7ħ&ëĦı
ėķĶ)½YĒÄėķĝģčďĶ.
Įĝ2Ĩ 3kHz MħşĽőŋŪíĒèņšŬŕħįĦH
XėķġčĶĒ, íħàėIJĜħ®į*ĸěřŏŮŬĹ»ē
ĕĶĖģĢ<¢ħņšŬŕĦHXėě, ďġŜūŞŨíħ?2
Ģħ9Ĺ_ęĶĖģĦı'ĵ®ĺĢčĔCĢČĶ.
[Lee 2009]ó Lee, A., Kawahara, T.: Recent development of open-source speech recognition engine JuliusűAPSIPAűpp. 131-137, 2009Ų
[Konomura ó 2013] ó Konomura, R., Hori, K.: Designing Hardware and Software Systems Toward Very Compact and Fully Autonomous QuadrotorsűIEEE/ASME International Conference on Advanced Intelligent Mechatronicsű2013Ų
[Konomuraó2014]ó Konomura, R., Hori, K.: Visual 3D Self Localization with 8 Gram Circuit Board for Very Compact and Fully Autonomous Unmanned Aerial Vehicles, 2014 IEEE International Conference on Robotics and Automation, 2014 (to appear)
[Monajjemi 2013]ó Monajjemi, V., et al: HRI in the sky: Creating and commanding teams of UAVs with a Vision-mediated Gestural Interface, Intelligent Robots and Systems (IROS), 2013 IEEE/RSJ International Conference on IEEE, pp. 617-623, 2013
[Ng 2011] Ng, W., et al: Collocated interaction with flying robots, RO-MAN/IEEE pp. 143-149, 2011
[Quigley 2004] Quigley, M., et al: Semi-autonomous human-UAV interfaces for fixed-wing mini-human-UAVs, IEEE/RSJ International Conference on IEEE, pp. 2457-2462, 2004
05. FIR śļŪŏů²:¥ãmĹđĕĝıħŰ
02. łŬŠŮ๥ń
04. ŜūŞŨíģşĽőŋŪíůŋŞńŔŨŢÁŰ
03. ŜūŞŨíůŋŞńŔŨŢÁŰ