I-Sensor Fusion algorithm

Isihlungi seKalman

Isihlungi seKalman sijwayelekile.

Umgogodla we-algorithm ukusetha iqoqo lezici "zokukholelwa" zenzwa ngayinye. Ngokwesikhashana ngasinye, idatha yezinzwa kusukela ngomzuzu wokugcina izosetshenziselwa izibalo ukuthuthukisa ukuqagela (ukuzingeza), futhi ikhwalithi yenzwa nayo izokwahlulelwa. Ngokuqhathanisa phakathi kwenani elibikezelwe kanye nenani elilinganisiwe lenzwa, inani elihle lizolinganiswa futhi likhiphe.

Lokhu kusho ukuthi uma inzwa ihlala inikeza inani elihle nelingaguquki futhi liqala ukukutshela okuthile okungathandeki, izinga lokutheleka sensor lizokwehla ngama-millisecond ambalwa aze aqale ngokwenza umqondo.

Lokhu kungcono kunokulinganisa okulula noma ukuvota, ngoba isihlungi se-kalman singabhekana nalesi simo esithi izinzwa eziningi ziphume okwesikhashana. Uma nje umuntu egcina isizathu esihle, angenza irobhothi ngomzuzu omnyama.

Kalman filter is an application of more general concepts of Markov chain and Bayesian reasoning, which is a mathematical system that iteratively improves their guesses by using evidence. Lawa mathuluzi angamathuluzi asetshenziselwa ukusiza isayensi ngokwayo imibono yokuhlola (okuyisisekelo salokho esikubiza ngokuthi "ukubaluleka kwezibalo").

Ngakho-ke, kungashiwo ngokwenkomba ukuthi izinhlelo ezithile ze-Sensor Fusion ziveza umongo wesayensi ngejubane lezikhathi eziyi-1000 ngomzuzwana.

Izihlungi zeKalman zisetshenzisiwe eziteshini ze-orbital zama-satellite amashumi eminyaka. Ngenxa yokuthi ama-miccocontroller anamuhla angasebenzisa i-algorithm ngesikhathi sangempela, aya ngokuya athandwa kakhulu kumarobhothi.