Sabuwar Hanyar Gano Motoci ta Haɗin Sadarwar Kamara ta Gani da Photogrammetry

Teburin Abubuwan Ciki

• 1. Gabatarwa
• 2. Tsarin Gine-ginen da Aka Gabatar
• 3. Hanyoyin Aiki
• 4. Sakamakon Gwaji
• 5. Bincike na Asali
• 6. Bayanan Fasaha da Tsarin Lissafi
• 7. Nazarin Shari'a: Yanayin Tafiya a Layi a Babbar Hanya
• 8. Aikace-aikace da Hanyoyi na Gaba
• 9. Manazarta

1. Gabatarwa

Gano wuri tsari ne na gano matsayi (x, y a sarari 2D; x, y, z a sarari 3D) na wani abu a wani takamaiman lokaci. Tare da haɓakar Intanet na Abubuwa (IoT) da motoci masu tafiya da kansu, gano wuri daidai ya zama mahimmanci. GPS na gargajiya yana ba da mafita na layin gani amma yana fama da matsalolin daidaito a cikin kwankwaso na birni da ramuka. Wannan takarda ta gabatar da sabuwar hanyar da ta haɗa Sadarwar Kamara ta Gani (OCC) da photogrammetry don cimma ingantaccen gano motoci ba tare da canza kayan aikin sufuri da ke akwai ba.

2. Tsarin Gine-ginen da Aka Gabatar

Tsarin yana rarraba motoci zuwa nau'i biyu: Motocin Mai Gida (HVs) waɗanda ke ƙididdige matsayin wasu motoci, da Motocin Mai Watsawa (FVs) waɗanda ke tafiya a gaban HVs. FVs suna watsa bayanan da aka canza daga fitilun bayansu, wanda kamarar HV ke karɓa ta amfani da OCC. Bugu da ƙari, ana amfani da bayanan fitilar titi (SL) don inganta daidaiton matsayin HV.

2.1 Muhimman Abubuwan Haɗin

• Sadarwar Kamara ta Gani (OCC): Yana amfani da haske mai canzawa daga fitilun bayan FV da SLs don watsa bayanai.
• Photogrammetry: Yana auna nisa ta hanyar ƙididdige yankin hoton da aka mamaye akan firikwensin hoto.
• Haɗa Bayanai: Yana haɗa bayanan OCC da photogrammetry don gano wuri mai ƙarfi.

3. Hanyoyin Aiki

HV yana ƙayyade matsayinsa ta amfani da bayanan SL, sannan yana ƙididdige matsayin dangi na FV ta hanyar kwatanta canje-canjen nisa tsakanin HV-SL da HV-FV. Ana ƙididdige nisa tsakanin FV ko SL da kamarar HV ta amfani da photogrammetry: $d = \frac{f \times H}{h}$, inda $f$ shine tsayin mai da hankali, $H$ shine ainihin tsayi, kuma $h$ shine tsayin hoto.

3.1 Ƙididdigar Nisa

Ta amfani da samfurin kamara mai rami, nisa $d$ daga kamara zuwa wani abu ana bayar da shi ta:

$d = \frac{f \times W}{w}$

inda $W$ shine ainihin faɗin abin kuma $w$ shine faɗin a cikin pixels akan firikwensin hoto.

3.2 Ƙididdigar Matsayi

An fara ƙididdige matsayin HV ta amfani da triangulation daga SLs da yawa. Sannan, ana ƙayyade matsayin dangi na FV ta:

$\Delta P_{FV} = P_{HV} + \Delta d \cdot \cos(\theta)$

inda $\Delta d$ shine canjin nisa kuma $\theta$ shine kusurwar isowa.

4. Sakamakon Gwaji

Saitin gwaji ya yi amfani da kamara mai ƙuduri 640x480, tsayin mai da hankali 3.6 mm, da fitilar baya mai diamita 0.15 m. Sakamako ya nuna kuskuren auna nisa ƙasa da 5% na nisa har zuwa mita 30. Tsarin da aka gabatar ya cimma daidaiton matsayi a cikin mita 0.5, wanda ya fi na mafita na GPS kawai wanda yawanci yana da kurakurai na mita 2-5.

5. Bincike na Asali

Mahimman Bayani: Wannan takarda ta gabatar da haɗe-haɗe na fasaha biyu da suka balaga—OCC da photogrammetry—don magance wata matsala mai mahimmanci a cikin tuki mai sarrafa kansa: gano motoci masu dogaro ba tare da haɓaka kayan aiki masu tsada ba. Babban sabon abu shine amfani da fitilun baya da fitilun titi da ke akwai a matsayin alamomin sadarwa, juya kayan aikin da ba su da aiki zuwa taimakon matsayi mai aiki.

Tsarin Tunani: Marubutan sun ci gaba da hankali daga gano matsala (iyakokin GPS) zuwa ƙirar mafita (OCC+photogrammetry), sannan zuwa ƙirar lissafi da tabbatarwa ta gwaji. Tsarin yana da daidaituwa, kodayake takarda za ta iya amfana daga kwatanta mai tsauri da hanyoyin zamani kamar SLAM na tushen LiDAR ko sadarwar V2X.

Ƙarfi da Rashi: Babban ƙarfi shine hanyar da ba ta da tsada, mai sauƙin kayan aiki. Duk da haka, tsarin yana ɗaukan layin gani mai kyau da yanayin haske mai kyau, wanda ba zai iya kasancewa a cikin hazo, ruwan sama, ko dare ba. Bugu da ƙari, dogaro ga canjin fitilar baya na iya shafar ta hanyar ƙazanta ko lalacewar fitilu. Idan aka kwatanta da tsarin tushen LiDAR (wanda ke da tsada dubunnan daloli), wannan hanyar ta tushen kamara ta fi arha amma ba ta da inganci a yanayi mara kyau. Kamar yadda Geiger et al. (2012) suka lura a cikin bayanan KITTI, hanyoyin tushen kamara sukan ragu a yanayin ƙarancin haske.

Shawarwari Masu Aiki: Ga masu aiki, wannan tsarin ya fi dacewa don tafiya a layi a babbar hanya da taimakon ajiye motoci inda yanayin haske ke sarrafawa. Aikin gaba ya kamata ya bincika hanyoyin haɗe-haɗe da ke haɗa OCC da radar ko firikwensin ultrasonic don aiki a kowane yanayi. Za a iya haɓaka samfurin photogrammetry na takarda ta amfani da ƙididdigar zurfin tushen koyon zurfi, kamar yadda Eigen et al. (2014) suka nuna a cikin aikinsu kan hasashen zurfin hoto guda ɗaya.

6. Bayanan Fasaha da Tsarin Lissafi

Samfurin photogrammetry yana amfani da lissafin kamara mai rami:

$\frac{x}{X} = \frac{f}{Z}$

inda $x$ shine haɗin hoto, $X$ shine haɗin duniya, $f$ shine tsayin mai da hankali, kuma $Z$ shine zurfi. Don sanannen girman abu $S$ da girman hoto $s$, nisa $D$ shine:

$D = \frac{f \times S}{s}$

Canjin OCC yana amfani da Maɓallin Kunnawa-Kashe (OOK) a mitoci sama da 100 Hz don guje wa kyalkyali da ake gani. Ana amfani da ƙarfin siginar da aka karɓa (RSS) don ƙididdigar nisa a matsayin hanya ta biyu:

$P_r = P_t \times \frac{A_r}{\pi D^2} \times \cos(\phi)$

inda $P_r$ shine ƙarfin da aka karɓa, $P_t$ shine ƙarfin da aka watsa, $A_r$ shine yankin mai karɓa, kuma $\phi$ shine kusurwar shiga.

7. Nazarin Shari'a: Yanayin Tafiya a Layi a Babbar Hanya

Yanayi: Rukunin motoci uku suna tafiya a gudun kilomita 80 a awa ɗaya a kan babbar hanya. Motar da ke gaba (FV) tana watsa saurinta da matsayin birki ta hanyar fitilun baya da aka canza. Motar tsakiya (HV) tana amfani da OCC don karɓar wannan bayanan da photogrammetry don auna nisa.

Matakan Aiwatarwa:

1. Fitilar baya ta FV tana canza bayanai a 200 Hz (OOK).
2. Kamarar HV tana ɗaukar hotuna a 30 fps, tana canza siginar.
3. Photogrammetry tana ƙididdige nisa: $D = \frac{3.6mm \times 0.15m}{h_{pixels} \times 0.006mm/pixel}$.
4. HV tana daidaita gudu don kiyaye amintaccen nisa (dokar dakika 2: ~44m a gudun kilomita 80 a awa ɗaya).
5. Idan FV ta yi birki, HV ta karɓi siginar a cikin 33 ms (hoto ɗaya) kuma ta mayar da martani.

Sakamako: Tsarin yana kiyaye tsarin rukunin motoci tare da daidaito na 0.5m, yana rage jan iska har zuwa 15% kuma yana inganta amfani da man fetur.

8. Aikace-aikace da Hanyoyi na Gaba

Tsarin da aka gabatar yana da aikace-aikace masu ban sha'awa da yawa na gaba:

• Ajiye Motoci Mai Sarrafa Kansa: Amfani da OCC daga fitilun filin ajiye motoci don sanya matsayi daidai.
• Gudanar da Mahadar Hanyoyi: Motoci suna sadarwa da fitilun zirga-zirga don inganta kwararar motoci.
• Gudanar da Rukunin Motoci: Bibiyar motocin isar da kaya a lokaci na gaskiya a yankunan birni.
• Haɗin V2X: Haɗa OCC da DSRC ko 5G don gano wuri mai maimaitawa.
• Kayan Aikin Birni Mai Wayo: Fitilun titi sun zama cibiyoyin sadarwa masu aiki da yawa.

Bincike na gaba ya kamata ya mai da hankali kan gano abubuwa na tushen koyon zurfi don inganta ƙarfi, da haɗin kai da firikwensin inertial don aiki mara tsinkaye yayin katsewar OCC.

9. Manazarta

1. M. T. Hossan et al., "A New Vehicle Localization Scheme based on Combined Optical Camera Communication and Photogrammetry," IEEE Access, 2021.
2. A. Geiger, P. Lenz, and R. Urtasun, "Are we ready for autonomous driving? The KITTI vision benchmark suite," CVPR, 2012.
3. D. Eigen, C. Puhrsch, and R. Fergus, "Depth map prediction from a single image using a multi-scale deep network," NeurIPS, 2014.
4. World Health Organization, "Global status report on road safety 2018," WHO, 2018.
5. J. Y. Kim et al., "Optical camera communication for vehicular applications: A survey," IEEE Communications Surveys & Tutorials, 2020.