201219789 六、發明說明： 【發明所屬之技術領域】 本發明是有關於一種車輛動作八 析程式（program)等，該車輛動作分^ ϋ車輛動作分 或在如險紅_雖未錢發生事故= 在事=發生時’ 時，對車輛的動作或周圍狀況等 ％成為事故 地對導致如上所述的狀況的原因進二:：算以可事後適當 【先前技術】 寻 車輛搭載型的行車記錄器例如自動’該 的”_部的影像進行記=== 出現險情科的客餘況、甚^生事故或 亦已出現如下的畅，即，例如i 查明=，搭載r上 照時間序列:逐二料收集裂置例如按 憶體（memory)内。接著，事後由其至記 體内的狀況資料進行參日s，# μ 〃 I置來+上述5己憶 分析（參照專利文獻减’可客觀地對事故等進行 然而，當事後欲對以上述方式記錄的狀況資料進行分 201219789 資料除了表示險情等的特定的車柄=:貝料，該狀況 已收集的多個狀況資料中，==抖:::類，自 況資料予以抽出。 /的表不車輛動作的狀 此處，先前是進行如下的作 地對已收集的多個狀況資料進即逐個 分類為碰撞事故、险—❿夺各狀況資料例如 以及單純的雜訊(1=險情無關的急刹車“ ^ 進行::的:ifΐ目Ϊ來逐個地對狀況資料進行確認且 的作業時間過長，而且難= 在士下_er)的隨忍的判斷來正確地進行分類。又，存 擔亦I 即，對使用者造成的肉體負擔以及精神負 另方面，如專利文獻2所示，可考虞如下的步晉， 了減輕使用者的負擔，自動地將狀況資齡類為 的狀況資料、與用以對險情等的動作進行分析 的波㈣2具體而言，上述裝置構成為基於加速度資料 抓/ >、波尚以及脈寬（pulsewidth)，將由雜訊引起的多 餘的狀況資料予以排除。 然而，若如上述裝置般，使用臨限值來對由雜訊引起 的夕餘的巧㈣、與収對險龄的動祕行分析的狀 、兄資料進行判別，則存在如下的問題，即，分類精度會因 201219789 u限值的没疋方法而大不相同。又，即便可使用臨限值來 將由雜訊引起的資料予以除去，亦必須藉由目視來逐個地 對剩餘的狀況資料進行確認且進行分類。 、、次例如’右將臨限值設定為低臨限值，則根據全部的狀 貝料’候補險情事例的抽出件數會增加，但抽出的資料 :的險料例的準僻變低，包含多餘的資料的比例變 祕面’若將臨限值設定為高臨限值，則根據全部 ’候補險情事例的抽出件數會減少，但抽出的 比例〜】、=險情事例的準確率變高，包含多餘的資料的 亦即&對於於該情形時’險情事例的遺漏率變高。 確率斑、賫、=臨限值來統一地進行分離的方法而言，準 臨㈣間存在取捨關係（tradeoff)。又，雖說使 形與藉由目視來進行分類的情 用= 二料來對險情事例進行分類，目前依然需要利 先前技術文獻 專利文獻 利特開2〇〇7_119°9號公報 【發明内i】日本彻42·號公報 著：下後， 每個動作’各加速度資料的特徵量‘== 5 「丨11 201219789， 本發明的主要的預期課題在於：並不僅 資，示的車輛的動作進行分類，而是自動地== 性來鐘定狀況資料所示的車輛動作。 ^了罪 狀、兄==發明的車姉作分析裝置的特徵在於包括： 狀况貝枓接收部，接收包含作用於車輛的前後加速度、左 下加速度的各加速度資料的狀況資料二或ί 1用於車輛的傾側角（_ angle)加速度、螺距角（_ 及偏航角（yawangie)加速度的各角加速度 =„料中的至少—種狀況資料；以及車輛動作鑑 。使用已接收的狀況資料中的前後加速度、左右加速 度的各加速度資料的特徵量的相對關係、或 声資螺距角加速度及偏航角加速度的各角加速 特徵量的相對關係中的至少—個特徵量的相對關 係，對上述狀況資料所示的車輛動作進行鑑定。 料的2旦可，各加速度資料的特徵量或各角加速度資 、阳ί 藉由上述特徵量的相對關係來默車輛動 、因此’可不依賴於目視而鑑定狀況資料所示的車輛動 二it不僅可將使用者的隨意的判斷予以排除而客觀 =作，而且可減輕使用者的時間負擔、肉體負 :^精神負擔。又，由於可根據各加速度資料的特徵量 2角加速度資料的特徵量的相對_來鑑定車輛動作， 與僅利舰限值來進行判別的情形相比較，可獲得 Τ罪性高的鑑定結果。 具體而言’本申請案發明人進行仔細研究之後，已知： 6 201219789 於以下（1)〜（3)的車輛動作中，前後加速度、左右加 速度以及上下加速度的各加速度資料的特徵量存在特有的 相對關係。亦即，當上述已接收的狀況資料包含前後加速 度、左右加速度以及上下加速度的各加速度資料時，考慮 由上述車輛動作鑑定部使用前後加速度、左右加速度以^ 上下加速度的各加速度資料的特徵量的相對關係，將上 狀況資料所示的車輛動作鑑定為以下的（1)〜（3) 任一種動作。 的 ⑴表示碰撞事故、與碰撞事故相關聯的險情或該險 情以外的急刹車的險情等動作。 等動表料祕駐崎料車輪以至姆的礙壓 (3)表示因兩輪通過車道上的凹凸而產生 (bound)的顛簸動作。 ’敗 更詳細而言，較佳為當前後加速度資料的特徵量大於 左右加速度貧料及上下加速度f料的特徵翻 =資料所示的車輛動作鑑定為^ It專動作’自上逑左右加速度資料 度資料及上下加速度資料的特 :：=加速 部將上述狀況資料所示的車輛、^ =動作鐘定 汁貼W咨粗糾-AA * ± 寻上返車輛動作鐘定部將上 述狀况紐所㈣車_作财為 丨肝上 對於上述險情等動作而言，為了更細^碰_ 201219789〆 故以及險情m卜的細車，較佳為上述狀況資料包含方向 燈資料（winker data)，該方向燈資料表示上述車輛的方向 燈的作動資訊，當鑑定為上述險料動作的狀況資料中包 含上述方向燈資料時’上述車輛動作較部將上述方向燈 資料所示的車輛動作，較為表示用以靠著車道的路肩停 又’本發明的車輛動作分析程式的特徵在於：使電腦 (computeO具有作為狀況資料接收部與車輛動作鑑定部 的功能，上述狀況資料接收部接收包含作用於車輛的前後 度加速度及上下加速度的各加速度資料的狀況 貧料、或匕g作用於車輛的傾側角加速度、螺距角加速度 及偏航角加速度的各角加速度資料的狀況資料巾的至少一 ，狀況資料’上述車輛動作鑑定部使用已接收的狀況資料 =前後加速度、左右加速度及上下加速度的各加速度資 料的特徵4_對_、或麵肖加速度、螺距角加速度 f偏航角加速度的各角加速度資料的特徵量的相對關係中 、至少—雜徵⑽相對難，對上述狀況資料所示 輛動作進行鑑定。 + j車補作分㈣置自搭餅車_行車記錄器取 :^料，對該狀況資料進行分析。亦可將上述車輛動 作刀析裝㈣功能解行車記制。亦即，本發明的行 $器的特徵在於包括：狀況資料接收部，接收包含作用 的前後加速度、左右加速度及上下加速度的各加速 又貝;的狀況資料、或包含作用於車輛的侧角加速度、 8 201219789. 螺距角加速度及偏航角加速度的各角加速度資料的狀況資 料中的至少一個狀況資料；以及車輛動作鑑定部，使用已 接收的狀況資料中的前後加速度、左右加速度及上下加速 度的各加速度資料的特徵量的相對關係、或傾侧角加速 度、螺距角加速度及偏航角加速度的各角加速度資料的特 徵1的相對關係中的至少-個特徵量的相對關係，對上述 狀況資料所示的車輛動作進行鑑定。 如此’於行輕_絲況:#料 =對各狀況資料所示的車輛動作進行鑑定，因，二 =對之後收集的狀況資料進行分_作業，或 。i二Γ擇特定的動作來儲存於行車記錄器的 。己匕體，從而可有效率地利用記憶體。 發明的效果 根據以上述方式構成的本發明， 而自動地且以高可靠性來鐘定狀 °下的方法 上述方法是指藉由目視來對狀;兄;作’ 資料進行判取進行分類。 “關如加速度 【實施方式】 (二=二=車㈣作分析系統 1.系統構成> 一⑽，對汽車(她的攝 nf 201219789 等，以及車輛動作分析裝置·，例如取得 器100所拍_動態影像資料等，基於該動態影 示的動態影像來較汽車v是否表現出規定的枓所 <2.行車記錄器> 特匕記錄器100是車輛搭載型的行車記錄器，該行車 ° 〇黏著於擋風玻璃（front glass)，或者設置 錶板（dashboard)附近或安裝於車輛内的適當的^於^ 故時或出現雖未致使事故發生卻有可能成為事 故的險情時’對前後固定編中的汽車V的動作或周 錄’上述行車記錄器100採用一體的構成，即， 將暴本構成要素即圖2所示的偵測單元3、資訊處理 8、報告單元4、輸人單元5、通信單元6、以及|脫= 錄單元7等収容於單個或多個外殼（casing)内。 ° 、偵測單元3對與汽車V的動作或周圍狀況等相關的狀 況進行感測（sensing)，且將表示該狀況的狀況資料予以 輸出，此處，至少使用有拍攝單元31、加速度感測器 (sensor ) 32、以及位置感測器33該3種裝置。拍攝單^ 31例如是電荷耦合元件（Charge Coupled Device，CCD) 相機（camera)，對自身車輛前方的車外狀況進行拍攝且 將表示車外狀況的圖像的狀況資料（動態影像資料）予以 輸出。加速度感測器32例如是利用壓電（piez〇)電阻效 應而構成的加速度感測器’該加速度感測器3 2對作用於^ 輛的三維的加速度進行感測，且將表示上述加速度的狀況 資料（加速度資料）予以輸出。具體而言，加速度感測器 201219789 32氧作用於車輛V的前後方向的前後加速度、作用於車輛 V的左右方向的左右加速度、以及作用於車輛v的上下方 向的上下加速度進行感測。位置感測器33例如為全球定位 系統（Global Positioning System，GPS)接收機，該 Gps 接收機接收（catch)來自多個衛星的電波而對車輛¥的位 置進行感測，且將表示該位置的狀況資料（位置資料）予 以輸出。再者，於狀況資料中，亦存在自車輛v的車速感 測器發送而來的車速資料、或表示車門（do〇r)的開閉的 車門開閉資料、或表示刹車的0N/0FF的刹車資料、以及 表示車輛V的方向燈的作動資訊的方向燈資料等，經由連 接器（connector) CN來接收上述資料。又，上述連接器 CN亦用作電源。 ° 報告單元4包含露出至外殼的表面的發光體即 LED41、或者内置於外殼的蜂鳴器（buzzer )或揚聲器 (speaker)等的聲音輸出體（未圖示）等。 。 此處，所謂輸入單元5，是指設置於外殼的表面的按 钮開關（button switch )。 此處，所謂通信單元6，是指内置於外殼，且與基站 或後述的車輛動作分析装置2〇〇之間收發電波的例如無線 區域網路（LocalAreaNetwork，LAN)或行動電話等的通 信用的硬體（hardware)。 此處，所謂裝脫式記錄單元7 ’是指可拔脫地安裴於 插槽（slot)的例如緊密快閃（c〇mpact Fiash，CF)記憶 卡（memorycard)或安全數位（SecureDigita卜 SD)記^ 11 201219789 if 卡’上述插槽於外殼的側方形成開口。 就構造而言，資訊處理單元8是包括中央處理單元 (Central Processing Unit ’ CPU) 81、内部記憶體 82 (例 如非揮發性記憶體）、以及I/O緩衝（buffer)電路（有時 亦包含類比數位（Analog Digital，AD)轉換器（converter) f) 83等的常說的電腦電路，該資訊處理單元8内置於外 忒。而且，上述CPU81根據儲存於記憶體82的規定區域 的程式來進行動作，#此，進行上述各單元的控制或資訊 處理。 簡單地進行說明，CPU81經常一面逐個地對行駛中的 各種狀況資料即加速度資料或位置資料、動態影像資料等 二面逐步將上^料暫時保存至設定於記憶體 、暫時（temp〇rary)區域（以下亦稱為暫時資料儲存 並且當產生如下的現象時，將該現象的前後固定期間 =的上述岐t鄉収記駐記龍82 己錄資料館存部)’上述現象間接地2 險if或事故、異常發生等的發生。 的^的加速度（減速度）超過規定基準值 逮度（減速度）的時間持續固定時間以上的 :於上述現r閉：形、以及車輛的電源關閉的情形等相 進仃讀魏，從而可儘可能地不記錄 12 201219789 無用的資料。 又，就防止記錄無用的資料的觀 錄器亦具有學習功能。亦即，於進打車記 r上述報告單元，藉由報告單元來 ==情紐生事故等，麟納來自駕駛人的“正# =以重複’藉此來某種程度地掌握駕駛人的駕駛傾= 例如’將加速度的規定基準值料 ' 示事故等㈣駕駛人所财的絲。 子S間接地表 p，根據已記錄陳況#料被記料況，對該 已圯錄的狀況資料進行加權’根據應記錄的重要八 類。接著’例如，當記憶體容量已滿時，自重要度^ 况資料起’練崎料料赚，接著記_雜況資料肖。 以上財式正規地被記錄的狀崎料則权場所 „送至分析中心（eenter)(未圖示），或者被移送至 農脫式記錄單元7，將該裝脫式記錄單元7取下，接 ^至分析中心’用於使用車輛動作分析震置200來進行事 後的分析。’ <3.車輛動作分析裝置> 車輛動作分析裝置200按照規定的各車輛動作 個搭載於汽車A的行車記錄器刚所獲得的狀況資 行分類，㈣支援事後的分析。上述車輛動作分析裝置 的具體的機械構成是包括CPU、記憶體、輸人輸出介面 (interface )、以及AD轉換器等的通用或專用的電腦，該 201219789 ^專用的電腦依照記憶於上述記憶體的規定區域的車 】3 i來分析程式，使cpu、周邊設備等協作，藉此，如 D示發揮作為狀況資料接收部201、狀況資料儲存 二、車輛動作鑑定部202、以及分析資料儲存部D2 的功能。 m ’使用圖4，一併對各部分m、20卜202、D2 各°卩分的動作進行說明。 資料接收部2〇1採納行車記錄器励的正規記錄 儲存的包含動態影像資料及各加速度資料 4㈣；4，將綠況#料儲存於狀況f料儲存部D1 (圖 況肢㈣接㈣2G1可包含接收狀 的通，?接收機’上述狀況資料是設置於行車記錄器觸 料，Ϊ = )6使用無線LAN等來發送的狀況資 ，况資科儲存部m儲存且蓄積著包含 s"2 (mz ；1 例如針對每ίί/π'τν狀況#料儲存部D1構成為 個狀況資料 ，系、·統儲存多輛汽車v所獲得的多 切的車ΐ兄動取得儲存於上述狀況資料儲存部 度、左右加速度以及上下加速产的的讀加速 ^ f Λ 又的各加迷度資料的特徵詈 進仃料（圖4、步驟S3)。本實施形__= 14 201219789.. 接著，車輛動作鑑定部202使用各加速度資料的 量即例如標準偏差的相對性大小_，來鑑定狀;兄資料ς :的車輛動作（® 4、步驟S4)。具體而言，車輛動作梦 定部202使用前後加速度、左右加速度以及上下加速产= 各加速度資料的標準偏差的相對性大小關係，將狀況J料 所不的車輛動作至少鑑定為以下的（1)〜（3)中的彳壬一 個動作。 U)表示碰撞事故、與碰撞事故相關聯的險情或該險 情以外的急刹車的險情等動作。 人双 々（2)表示單輪碾壓至路緣石或車輪脫入至側溝的碾壓 等動作。 (3)表示因兩輪通過車道上的凹凸而產生的顛簸 簸動作。 更詳細而言，如圖5所示，當前後加速度資料的加速 度波形的標準偏差，大於左右加速度資料的加速度波形的 標準偏差及上下加速度資料的加速度波形的標準偏差時， 車輛動作鑑定部202將上述狀況資料所示的車輛動作鑑定 為上述險情等動作。原因在於：由於碰撞時的衝擊或急刹 車4的衝擊，各加速度中的前後加速度會最大幅度地發生 變動，從而前後加速度波形的不均變得最大。再者，圖5 是模式圖，藉由目視來將上述行車記錄器1〇〇所獲得的多 15 201219789 個狀況資料分類為上述（1)〜（3)，針對上述每個動作而 對狀況資料進行排列，表示各狀況資料的各加速度波形的 標準偏差。圖5的橫軸表示各狀況資料’縱軸表示各加速 度波形的標準偏差。 又，當上述左右加速度資料的加速度波形的標準偏 差，大於前後加速度資料的加速度波形的標準偏差及上下 加速度資料的加速度波形的標準偏差時，車輛動作鑑定部 202將上述狀況資料所示的車輛動作鑑定為上述碾壓等動 =(參照圖5)。原因在於：在碾壓等動作中，僅車輛卩的 刖輪中的單輪會急遽地發生變動，藉此，各加速度中的左 右加速度會最大幅度地發生變動，從而左右加速度波 不均變得最大。 :且，吾上下加速度資料的加速度波形的標準偏差， 大於前後加速度資料的加速度波形的標準偏差及上述左 加速度資料的加速度波形的標準偏差時，車 = 況資料所示的車輛動作蚊為上述顛疑血^ 夂Μη 動Ρ大致同時地上下運動，藉此， 下力下加速度會最大幅度地發生變動，從而上 卜加迷度波形的不均變得最大。 :且，當鑑定為上述險情轉作的狀況資料 ΓΓ車輛動作鐘定部202將上述方向燈資料 魚刹ΐίίΐ 定為表示用以靠著車道的路肩停車‘ 一、動作（路肩停止動作）（參照圖5)。以上述方式 16 201219789 =者藉:圖rrr更細化地對險情等動作進行 向燈的有無。又，利二2的7=向細線來表示方 的密集部分。】用〜線㈤chmg)來表示方向燈信號 定資料與以202使作為鑑定結果的動作鑑 聯，將該動作3=，行動作鑑定的狀況資料相關 步驟S5) 抖儲存於分析資料儲存部D2 (圖4、 作已的動作’系統地對動 具體而言行分類且進行儲存。 動作存部D2將相對應的狀況資料以及 及顛錄純’攝針對險情等動作、賴等動作、以 i==r疋的儲存資料卿e〇内。例如於分析 = 設定錯料祕資、碾壓等動作資 =二=:==:動作資料。 =著―壓=二== 内，儲存有姆_狀崎料以及動作鑑定各貝枓失 而^分析貧料儲存部D2對上述險情等動作資料夾 ，從而狀路肩停止動作資料夾，該路 月分止動作貝料爽用以自險情等 肩停止動作且儲存著該路肩停止動作進7刀類出上述路 17 201219789〆 操作員（operator )對例如鍵盤（key board )或滑鼠 (_Se)等的輸人單㈣行操作，藉此，僅選擇以上述 方式儲存於分析資簡存部Μ陳況#料巾的被分類為 特定的車輛動作的狀況資料，將該狀況·輸出至例如顯 不器（display)等的輸出單元。或者，同樣地僅選擇被分 類為特定的車姉作的狀況資料，將錄況資料傳送至其 他分析裝置或記憶體等。 〃 <本實施形態的效果> 根據以上述方式構成的本實施形態的車輛動作分析系 統，可根據前後加速度資料、左右加速度資料及上下加速 度資料所示的前後加速度波形的標準偏差、左右加速度波 形的標準偏差及上下加速度波形的標準偏差的大小關係， 來鑑定車輛動作。因此，可不依賴於目視而鑑定狀況資料 所示的車輛動作，從而不僅可將使用者的隨意的判斷予以 排除而客觀地鑑定車輛動作，而且可減輕使用者的時間負 擔、肉體負擔以及精神負擔。又’可不使用臨限值而根據 前後加速度波形的標準偏差、左右加速度波形的標準偏差 以及上下加速度波形的標準偏差的大小關係來鑑定車輛動 作’因此’可獲得可靠性高的鑑定結果。 <其他變形實施形態> 再者，本發明並不限於上述實施形態。 例如，於上述實施形態中使用有各加速度資料，但若 行車記錄器100包括陀螺儀感測器（gyrosensor)，則亦可 使用作用於車輛的傾側角加速度、螺距角加速度以及偏航 201219789 角加速度的各肖加速度資料㈣ 又’於上述實施形態中，分柄輛動作。 情等動作、賴賴作、齡部包括針對險 J 定動作’但不限於此。例如，車 抽出;儲括狀況資料抽出部，該狀況資料 ，疋的車輛動作的狀況資料予以抽出。而且，狀況資= 出^亦可基於賦錄況資料的動作鑑定龍，將如下 況資料予以抽出，該狀況資料是由择 / 鼠專的輸人早⑽行操作來選擇，且表示規定的車輛動作。 而^於上述實施形態中’將行車記錄器⑽所獲得 ^大況貝料收集至車輛動作分析裝置之後，藉由該車輛動 作分析裝絲鑑定狀況資料所福車輛動作，但於行車纪 錄器励中，亦同樣可鑑定狀況資料所示的車柄動作。例 =，打車記錄器100包括狀況資料接收部以及車輛動作鑑 定部’亦可僅將由上述車輛動作鑑定部鑑定為蚊的動^ (例如險情_作）離況資料，料且轉至記憶體内 的正規區域（正規記錄資料儲存部）。 而且，亦可使用規定的臨限值，將被分類為險情等動 作的狀況資料予以細化。例如，可考慮根據如下的值與臨 限值的關係，將上述狀況資料予以細化，上述值是自加速 度感測器所獲得的3軸合成加速度的最大值減去上下加速 度最大值所得的值。 又，為了更細化地鑑定碰撞事故，較佳為將險情等動 19 201219789f201219789 VI. Description of the Invention: [Technical Field] The present invention relates to a vehicle operation program, etc., the vehicle action is divided into a vehicle action or in a dangerous red _ although there is no money accident = When the event occurs, the vehicle's operation or surrounding conditions become an accident. The cause of the above-mentioned situation is two:: It can be calculated afterwards. [Prior Art] Looking for a vehicle-mounted driving recorder For example, the image of the automatic 'this' part is recorded === The customer's condition of the dangerous situation, the accident or the following has also appeared, that is, for example, i found =, equipped with r time series: The material is collected for two times, for example, within the memory. Then, afterwards, the status data from the body to the body is used to perform the s day s, # μ 〃 I set + the above 5 recall analysis (refer to the patent document minus ' It is possible to objectively perform an accident, etc. However, in the event that the situation data recorded in the above manner is to be subdivided in 201219789, in addition to the specific handle indicating the danger, etc., the material is collected, and the situation has been collected. =Shake::: Class, self-sufficient data is extracted. / The form of the vehicle does not move here. Previously, the following conditions were used to classify the collected multiple status data into collision accidents, risks - usurpment of various status data, for example. And simple noise (1 = sudden brakes that are not related to danger) ^ Conducted::: If you look at the status data one by one, and the working time is too long, and it is difficult = in the _er) The judgment is made to classify correctly. In addition, the burden is also the same, that is, the physical burden on the user and the mental burden, as shown in Patent Document 2, the following steps can be taken to reduce the burden on the user. Specifically, the status data of the status-age class and the wave (4) for analyzing the action such as the danger are automatically configured. Specifically, the device is configured to capture based on the acceleration data, >, and the pulse width and the pulse width. Exclude the excess situation data caused by noise. However, if the device is used as described above, the threshold value is used to analyze the sensation (4) caused by the noise, and the analysis of the secrets of the aging age. Brother data Judging, there is a problem that the classification accuracy is greatly different due to the flawless method of the 201219789 u limit. Moreover, even if the threshold value can be used to remove the data caused by the noise, it is necessary to visually The remaining status data are confirmed and classified one by one. For example, if the right threshold value is set to the low threshold value, the number of extractions of the candidate condition will increase according to the total shape, but the number of extractions will increase. The extracted data: the probability of the dangerous case is low, and the proportion of the excess data is changed to the secret surface. If the threshold is set to the high threshold, the number of extractions will be reduced according to the case of all the candidates. However, the ratio of the extracted ratios ~], = the case of the dangerous case becomes high, and the data containing the excess information is & for this case, the missing rate of the dangerous case becomes higher. In the method of determining the uniformity of the spot, the 賫, and the threshold value, there is a tradeoff between the four (4). In addition, although the classification and the classification by visual inspection = the second material to classify the dangerous cases, it is still necessary to obtain the prior art literature patent document Lite Open 2〇〇7_119°9 bulletin [invention i] Japan's Toru 42. bulletin: After the next, each action 'feature quantity of each acceleration data' == 5 "丨11 201219789, the main expected subject of the present invention is: not only the capital, the movement of the vehicle is classified Instead, it automatically == sex to determine the vehicle action shown in the status data. ^The guilt, brother == the invention of the rut analysis device is characterized by: the condition of the beijing receiving unit, the receiving includes the action on the vehicle The front and rear accelerations, the condition data of the acceleration data of the lower left acceleration, or the ί 1 for the vehicle's tilt angle (_ angle) acceleration, the pitch angle (_ and the angular acceleration of the yawie acceleration = ??? At least the status data; and the vehicle action check. The relative relationship between the feature quantities of the acceleration data of the front-back acceleration and the left-right acceleration in the received status data, or the sound snail The relative relationship of at least one feature quantity in the relative relationship between the angular acceleration and the angular acceleration characteristic of the yaw angle acceleration is identified by the vehicle behavior indicated by the above condition data. The characteristics of each acceleration data are 2 The quantity or the angular acceleration and the yang are used to illuminate the vehicle by the relative relationship of the above-mentioned characteristic quantities, so that the vehicle movement shown in the identification status data can be excluded from the user's arbitrary judgment without relying on the visual observation. Objectively = work, and can reduce the user's time burden, physical negative: ^ mental burden. In addition, because the feature quantity of each acceleration data can be used to identify the vehicle's movements, In comparison with the case where the ship limit is used for discrimination, an identification result with high sinfulness can be obtained. Specifically, after the inventors of the present application conducted a careful study, it is known that: 6 201219789 is as follows (1) to (3) In the vehicle operation, there is a unique relative relationship between the characteristic quantities of the acceleration data of the front-rear acceleration, the left-right acceleration, and the vertical acceleration. That is, when When the received condition data includes the acceleration data of the front-rear acceleration, the left-right acceleration, and the vertical acceleration, the relative relationship between the characteristic quantities of the acceleration data and the acceleration acceleration of the left and right accelerations by the vehicle motion identification unit is considered. The vehicle behavior indicated by the status data is identified as any of the following (1) to (3) operations. (1) indicates a collision accident, a danger associated with the collision accident, or a dangerous situation other than the dangerous situation. The material impediment of the car's wheel and the pressure of the hammer (3) indicates the bumping action caused by the two rounds of the bump on the lane. 'In more detail, it is better that the feature quantity of the current post-acceleration data is greater than The characteristics of the left and right accelerations and the acceleration of the upper and lower accelerations are as follows: The motion of the vehicle indicated by the data is identified as ^ It special action 'Special data from the upper and lower acceleration data and the upper and lower acceleration data::= Acceleration part shows the above situation data Vehicle, ^ = action bell juice paste W consultation rough correction - AA * ± find the return vehicle action clock will be the above situation (4) _Fortunately for the above-mentioned dangers and other actions in the liver, in order to more carefully touch the _ 201219789 and the dangerous car, it is better to include the direction data (winker data), the direction light The data indicates the operation information of the directional light of the vehicle, and when the status data identifying the dangerous material operation includes the directional light data, the vehicle movement comparison unit operates the vehicle indicated by the directional light data, and indicates that the vehicle is operated. The vehicle motion analysis program of the present invention is characterized in that the computer (computeO has a function as a status data receiving unit and a vehicle motion identifying unit, and the status data receiving unit receives the acceleration before and after acting on the vehicle) And at least one of the condition data sheets of the acceleration data of the vertical acceleration or the acceleration of the vehicle, or at least one of the angular acceleration data of the angular acceleration of the vehicle, the pitch angular acceleration, and the yaw angular acceleration. The appraisal department uses the received status data = front and rear acceleration, left and right acceleration, and up and down The relative relationship between the characteristics of each acceleration data of 4_ _, or surface acceleration, pitch angular acceleration f yaw acceleration, and at least the miscellaneous sign (10) is relatively difficult. The action shown is identified. + j car make up points (four) set from the cake car _ driving recorder to take : ^ material, the analysis of the status data. It is also possible to disassemble (4) the above-mentioned vehicle motion tool to record the vehicle. That is, the line device of the present invention is characterized in that the condition data receiving unit receives the condition data including the front and rear accelerations of the action, the left and right accelerations, and the accelerations of the vertical acceleration, or includes the side angular acceleration acting on the vehicle. , 8 201219789. At least one condition data in the condition data of each angular acceleration data of the pitch angular acceleration and the yaw angular acceleration; and the vehicle motion identification unit, using the longitudinal acceleration, the left and right acceleration, and the vertical acceleration in the received condition data The relative relationship between the feature quantities of each acceleration data, or the relative relationship of at least one feature quantity in the relative relationship of the feature 1 of each angular acceleration data of the tilt angular acceleration, the pitch angular acceleration, and the yaw angular acceleration, The vehicle behavior shown is identified. Such a 'light line _ silk condition: # material = identification of the vehicle action shown in each status data, because, 2 = the status data collected after the division _ homework, or. i choose a specific action to store on the driving recorder. It is a body that can use memory efficiently. Advantageous Effects of Invention According to the present invention configured as described above, the method of automatically and highly reliablely determining the above-described method refers to sorting by visual observation; brother; "Off as acceleration [Embodiment] (2 = 2 = car (4) for analysis system 1. System configuration > 1 (10), for the car (her photo nf 201219789, etc., and vehicle motion analysis device, such as the taker 100 _Dynamic image data, etc., based on the dynamic image of the motion picture, whether or not the vehicle v exhibits a predetermined position. <2. Driving recorder> The recorder 100 is a vehicle-mounted driving recorder, and the vehicle is driven. ° 〇 〇 〇 front front front , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , front , front front front front front front front The operation of the vehicle V in the front and rear fixed knitting or the recording of the above-mentioned driving recorder 100 is an integral configuration, that is, the detecting unit 3, the information processing 8, the reporting unit 4, and the input shown in FIG. The human unit 5, the communication unit 6, and the |re-recording unit 7 are housed in a single or a plurality of casings. ° The detecting unit 3 senses conditions related to the operation of the vehicle V or the surrounding conditions. (sensing), and The status data indicating the status is outputted. Here, at least three types of devices, the imaging unit 31, the acceleration sensor 32, and the position sensor 33 are used. The photographing unit 31 is, for example, a charge coupled device (Charge) Coupled Device, CCD) A camera that captures the outside situation of the vehicle in front of the vehicle and outputs status information (moving image data) of the image indicating the situation outside the vehicle. The acceleration sensor 32 uses, for example, piezoelectric (piez)加速度) an acceleration sensor composed of a resistance effect. The acceleration sensor 3 2 senses a three-dimensional acceleration acting on the vehicle, and outputs status information (acceleration data) indicating the acceleration. Specifically, The acceleration sensor 201219789 32 oxygen is applied to the longitudinal acceleration in the front-rear direction of the vehicle V, the left-right acceleration acting in the left-right direction of the vehicle V, and the vertical acceleration acting in the vertical direction of the vehicle v. The position sensor 33 is sensed. For example, a Global Positioning System (GPS) receiver that catches from a GPS receiver The position of the vehicle ¥ is sensed by the radio waves of the satellites, and the status data (location data) indicating the position is output. Furthermore, in the status data, there is also a vehicle speed sensor transmitted from the vehicle v. The vehicle speed data, or the door opening and closing data indicating the opening and closing of the door (or 〇r), the brake data indicating the brake 0N/0FF, and the direction lamp data indicating the operation information of the direction light of the vehicle V, etc., via the connector ( Connector) CN to receive the above information. Further, the above connector CN is also used as a power source. The reporting unit 4 includes an LED 41 that is an illuminant exposed to the surface of the casing, or a sound output body (not shown) such as a buzzer or a speaker built in the casing. . Here, the input unit 5 refers to a button switch provided on the surface of the casing. Here, the communication unit 6 is a communication device such as a wireless area network (Local Area Network) or a mobile phone that is built in the casing and transmits and receives radio waves between the base station or the vehicle motion analysis device 2 described later. Hardware. Here, the detachable recording unit 7' refers to, for example, a compact flash (c〇mpact Fiash, CF) memory card or a secure digital (SecureDigitab SD) that is detachably mounted in a slot. ) ^ 11 201219789 if card 'The above slot forms an opening on the side of the outer casing. In terms of structure, the information processing unit 8 includes a central processing unit (CPU) 81, internal memory 82 (for example, non-volatile memory), and an I/O buffer circuit (sometimes also included Analog Digital (AD) converter (converter) f) 83 commonly used computer circuit, the information processing unit 8 is built in the outer casing. Further, the CPU 81 operates in accordance with a program stored in a predetermined area of the memory 82, thereby performing control or information processing of each unit described above. Briefly, the CPU 81 temporarily saves the materials on the two sides of the various situation data, that is, the acceleration data, the position data, and the motion image data, one by one, to the memory, temporary (temp〇rary) area. (The following is also referred to as temporary data storage and when the following phenomenon occurs, the above-mentioned fixed period of the phenomenon = the above-mentioned 乡t township recorded in the record of the dragon 82 recorded in the library)) the above phenomenon indirectly 2 insurance if or Accidents, abnormalities, etc. occur. The acceleration (deceleration) of ^ exceeds the predetermined reference value (deceleration) for a fixed time or longer: in the above-mentioned r closed: shape, and the power supply of the vehicle is turned off, etc. As far as possible, do not record 12 201219789 useless data. Also, the viewer that prevents the recording of useless data has a learning function. That is to say, in the above-mentioned report unit of the taxi, the report unit is used to == love the accident, etc., and Linna comes from the driver's "positive #=repetition" to somehow grasp the driver's Driving tilt = for example, 'the specified reference value of the acceleration' indicates the accident, etc. (4) The silk of the driver's money. The sub-S indirectly, the table p, according to the recorded condition, the recorded status information The weighting is based on the important eight categories that should be recorded. Then, for example, when the memory capacity is full, the data from the importance data is used to make a profit, and then the _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ The recorded esoteric material is sent to an analysis center (not shown) or transferred to the agricultural record unit 7, and the detachable recording unit 7 is removed and connected to the analysis center. 'Used to analyze the shock 200 using the vehicle motion analysis for post-mortem analysis. <3. Vehicle motion analysis device> The vehicle motion analysis device 200 classifies the status of the traffic recorder newly installed in the vehicle A in accordance with the predetermined vehicle operation, and (4) supports the subsequent analysis. The specific mechanical configuration of the vehicle motion analysis device described above is a general-purpose or dedicated computer including a CPU, a memory, an input interface, and an AD converter. The 201219789^ dedicated computer is stored in the memory. In the vehicle of the predetermined area, the program is analyzed, and the cpu, the peripheral device, and the like are collaborated. The D data is used as the status data receiving unit 201, the status data storage unit 2, the vehicle behavior identifying unit 202, and the analysis data storage unit D2. The function. The operation of m ′ is described with reference to Fig. 4 and the division of each of the parts m, 20, 202, and D2. The data receiving unit 2〇1 adopts the normal recording and storage of the driving recorder to include the moving image data and the acceleration data 4 (4); 4, the green condition # material is stored in the condition f material storage unit D1 (Fig. 4 (4) 4G1 can include In the receiving state, the receiver's status data is set to the driving recorder, and Ϊ = ) 6 is transmitted using a wireless LAN or the like. The conditional storage unit m stores and stores s"2 ( Mz ;1 For example, for each ίί/π'τν condition # material storage unit D1 is configured as a status data, and the system is stored in the above-mentioned status data storage unit by storing a plurality of cars v obtained by the system. The degree of acceleration, the acceleration of the left and right acceleration, and the acceleration of the acceleration and acceleration of the upper and lower accelerations are further improved (Fig. 4, step S3). The present embodiment is __= 14 201219789.. The identification unit 202 uses the amount of each acceleration data, that is, the relative magnitude _ of the standard deviation, for example, to identify the vehicle behavior of the brother data ® : (4, step S4). Specifically, before and after the vehicle motion dreaming unit 202 is used Acceleration, left and right acceleration And up and down acceleration = the relative magnitude relationship of the standard deviation of each acceleration data, and at least one of the following (1) to (3) is identified as the vehicle movement in the case of the condition J. U) indicates the collision accident The danger associated with a collision accident or the danger of sudden braking other than the dangerous situation. The double 々 (2) indicates the action of rolling a single wheel to the curb or the rolling of the wheel into the side groove. (3) indicates a bumpy movement caused by the unevenness of the two lanes passing through the lane. More specifically, as shown in FIG. 5, when the standard deviation of the acceleration waveform of the current post-acceleration data is greater than the standard deviation of the acceleration waveform of the left and right acceleration data and the standard deviation of the acceleration waveform of the upper and lower acceleration data, the vehicle motion identifying unit 202 The vehicle behavior indicated by the above situation data is identified as the above-mentioned danger and the like. The reason is that the longitudinal acceleration of each acceleration changes greatly due to the impact at the time of collision or the impact of the emergency brake 4, and the unevenness of the longitudinal acceleration waveform becomes maximum. In addition, FIG. 5 is a schematic diagram in which the plurality of 201219789 status data obtained by the above-described driving recorder 1A are classified into the above (1) to (3) by visual observation, and the status data is described for each of the above actions. The arrangement is performed to indicate the standard deviation of each acceleration waveform of each status data. The horizontal axis of Fig. 5 indicates the respective status data. The vertical axis indicates the standard deviation of each acceleration waveform. Further, when the standard deviation of the acceleration waveform of the left and right acceleration data is greater than the standard deviation of the acceleration waveform of the longitudinal acceleration data and the standard deviation of the acceleration waveform of the vertical acceleration data, the vehicle behavior identification unit 202 operates the vehicle indicated by the situation data. It was identified as the above-mentioned rolling press etc. (refer to FIG. 5). The reason is that in the operation such as rolling, only the single wheel in the wheel of the vehicle 遽 is violently changed, whereby the left and right accelerations in the respective accelerations are changed to the maximum extent, and the left and right acceleration waves become uneven. maximum. : Moreover, when the standard deviation of the acceleration waveform of the acceleration data is greater than the standard deviation of the acceleration waveform of the longitudinal acceleration data and the standard deviation of the acceleration waveform of the left acceleration data, the vehicle motion mosquito shown by the vehicle = condition data is the above-mentioned Suspected blood ^ 夂Μ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ And, when it is identified as the situation data of the above-mentioned dangerous situation, the vehicle operation timing unit 202 sets the above-mentioned directional light data fish brake ΐ ΐ ΐ 为 为 为 为 表示 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一Figure 5). In the above way 16 201219789 = Borrowing: Figure rrr more detailed on the danger and other actions to the presence or absence of lights. Further, 7= of the 2nd 2 is a dense part of the square. 】 Use the line (five) chmg) to indicate the direction lamp signal setting data and 202 to identify the action as the identification result, and the action 3=, the line item identification status data related step S5) is stored in the analysis data storage unit D2 ( Figure 4, the action has been systematically classified and stored in the system. The action memory D2 will be the corresponding status data and the automatic recording of the action, such as the danger, etc., with i== r疋 The storage information is in the e. For example, in the analysis = setting the wrong material secrets, crushing, etc. = two =: ==: action data. = "pressure = two == inside, stored with _ _ The raw material and the motion identification of each of the shellfish are lost. ^Analysis of the poor material storage unit D2 on the above-mentioned dangerous situation and other action data folders, so that the road shoulder stops the action folder, and the road is divided into the end of the action. And storing the shoulder stop action into the 7-knife out of the above-mentioned way 17 201219789 〆 operator (operator) to the input (four) line operation such as a keyboard (key board) or a mouse (_Se), thereby selecting only The above method is stored in the analysis of the assets and savings department. The status data classified into the specific vehicle operation is output to the output unit such as a display, etc. Alternatively, only the status data classified as the specific vehicle is selected, and the status information is recorded. The data is transmitted to another analysis device, a memory, etc. 〃 <Effects of the present embodiment> According to the vehicle motion analysis system of the present embodiment configured as described above, the longitudinal acceleration data, the left and right acceleration data, and the vertical acceleration data can be used. The standard deviation of the longitudinal acceleration waveform, the standard deviation of the left and right acceleration waveforms, and the standard deviation of the vertical acceleration waveform are used to identify the vehicle motion. Therefore, it is possible to identify the vehicle motion indicated by the situation data without depending on the visual observation. The user's arbitrary judgment can be excluded to objectively identify the vehicle movement, and the time burden, the physical burden, and the mental burden of the user can be alleviated. Moreover, the standard deviation and the left and right acceleration of the longitudinal acceleration waveform can be used without using the threshold value. Standard deviation of the waveform and acceleration up and down The magnitude relationship of the standard deviation of the waveforms is used to identify the vehicle behavior, so that a highly reliable identification result can be obtained. <Other variant embodiments> Further, the present invention is not limited to the above embodiment. For example, in the above embodiment Various acceleration data are used, but if the driving recorder 100 includes a gyro sensor, it is also possible to use the angular acceleration of the vehicle, the angular angular acceleration, and the angular acceleration of the yaw 201219789 angular acceleration (4). In the above-described embodiment, the operation of the split handle is performed. The action such as the situation, the operation, and the age include the action of the danger, but the present invention is not limited thereto. For example, the vehicle is taken out; the status data extraction unit is included, and the status information is The situation information of the vehicle movement is extracted. Moreover, the status of the capital = output ^ can also be based on the action of the information on the status of the identification of the dragon, the following information is extracted, the status data is selected by the selection of the mouse / mouse special (10) line operation, and indicates the specified vehicle action. In the above embodiment, after the vehicle condition recorder (10) obtains the conditional material and collects it into the vehicle motion analysis device, the vehicle behavior is analyzed by the vehicle motion to identify the situation data, but the vehicle is motivated by the driving recorder. In the same way, the handle movement indicated by the situation data can also be identified. For example, the taxi recorder 100 includes a status data receiving unit and a vehicle motion identifying unit ′, and may only use the vehicle motion identifying unit to identify the moving information of the mosquito (for example, danger), and transfer the data to the memory. Regular area within (regular record data storage). Furthermore, it is also possible to use the prescribed threshold to refine the situation data classified as dangerous. For example, it is conceivable to refine the above-mentioned situation data according to the relationship between the value and the threshold value, which is a value obtained by subtracting the maximum value of the vertical acceleration from the acceleration of the 3-axis synthetic acceleration obtained by the acceleration sensor. . Moreover, in order to more accurately identify the collision accident, it is better to be in danger. 19 201219789f
t t JIX 作鐘定為表示如下的值的動作，該值的左右加速度資料、 =力及前後加速度資料的特徵量，顯著地大於 力的路面與輪胎（响）的接地面的摩擦 t、車輛動作分析裝置亦可使狀況資料與地圖資訊 ϋ )。具體而言’車輛動作分析裝置使用狀況資料 所έ的位置資料，將地圖資訊與狀況資料予以鏈結。藉 此’例如可根據被鐘定為麵動作的狀況資料來推測出道 路資訊（例如路面的劣化狀態等），從而進行道路分析。 此外，亦可使用前後加速度資料的變動形態，對碰撞 事故中的與其他車輛或構造物等發生碰撞時的動作、及與 ，他車輛發生碰撞的動作進行分類。亦即’對於上述險^ ，動作而言，為了鑑定追尾事故或被追尾事故，較理想的 是根據左右加速度資料、上下加速度資料及前後加速度資 料的特徵量的正負來進行鑑定。 而且，除了上述實施形態之外，亦可使用表示礙壓等 動作的狀況資料中所含的方向燈資料，以如下的方式來對 該狀況資料進行處理。亦即，亦可根據方向燈資料來對車 輛的左拐或右拐進行判別，例如，確認在車輛V左拐時或 右拐時是否已碾壓至路緣石等。 又，當判斷是已右拐且已碾壓至路緣石，還是已左拐 且已碾壓至路緣石時’無方向燈資料’對左右加速度的波 形的正負的順序進行觀察’藉此，可分清左右的輪胎的碾 壓順序，因此，可進行判斷。 201219789 藉，可谷易地對駕駛人的駕駛傾向進行 列如是指在左拐時容易碾麗至路緣石1 在右拐時容易碾壓至路緣石。 、 或 u 皮形進行輪來判斷出 7轉考（turn)，因此，例如亦可判斷 轉彎時的事故或險情。 否為進仃U形 ϋΐΐ實施形S的加速射料的特徵量為加速度波形的 ‘準偏差’但可使用其他加速度波形的變異數（_ 亦可使用加速度的平均值。料，對於各加速度，亦可 用表示各動作特有的大小關係的運算值。 而且，本發明並不限於上述實施形態，當然可於 離本發明的宗旨的範圍内進行各種變形。 產業上的可利用性 根據如上所述的構成的本發明，並不僅藉由目視來 狀況資料所示的車輛的動作進行分類，而是可自動地且以 尚可罪性來鑑定狀況資料所示的車輛動作。 【圖式簡單說明】 圖1是模式性地表示本實施形態的車輛動作分析系 的圖。 '' 圖2是表示本實施形態的行車記錄器的構成要素的 圖。 、 圖 圖3是本實施形態的車輛動作分析裝置的功能構成 圖4是表示本實施形態的車輛動作分析裝置的動作的 21 201219789f 流程圖。 圖5是表示本實施形態中的車輛動作與各加速度的特 徵量的對應關係的圖。 【主要元件符號說明】 3:偵測單元 4:報告單元 5:輸入單元 6 :通信單元（發送機） 7:裝脫式記錄單元 8:資訊處理單元 31 :拍攝單元 32 :加速度感測器 33:位置感測器 41 :發光體（LED)Tt JIX is determined as the action of the following values. The left and right acceleration data, the force of the value and the characteristic amount of the longitudinal acceleration data are significantly larger than the friction between the ground surface of the force and the tire (ring). The analysis device can also make status data and map information ϋ). Specifically, the location information of the vehicle motion analysis device usage status data links the map information and the status data. By this, for example, it is possible to estimate road information (e.g., deterioration state of the road surface) based on the situation data that is determined to be a surface motion, thereby performing road analysis. In addition, it is also possible to use a change pattern of the longitudinal acceleration data to classify an action when a collision occurs with another vehicle or a structure, and an operation in which the vehicle collides. That is, in order to identify the rear-end accident or the rear-end collision for the above-mentioned dangers and actions, it is preferable to perform the identification based on the positive and negative characteristics of the left and right acceleration data, the vertical acceleration data, and the front and rear acceleration data. Further, in addition to the above-described embodiment, the directional lamp data included in the status data indicating the operation such as the squeezing pressure may be used to process the status data as follows. That is, it is also possible to discriminate the left or right turn of the vehicle based on the direction lamp data, for example, to confirm whether the curb has been crushed to the curb when the vehicle V turns left or turns right. Also, when it is judged that it has been turned right and has been crushed to the curb, or has been turned left and has been crushed to the curb, the 'non-directional light data' observes the positive and negative sequence of the waveform of the left and right accelerations. The order of rolling of the left and right tires is distinguished, and therefore, the judgment can be made. 201219789 Borrowing, you can easily drive the driver's driving tendency. If you turn left, it is easy to run to the curb. 1 It is easy to crush to the curb when turning right. , or u The leather shape is used to determine the 7 turn. Therefore, for example, it is also possible to judge the accident or danger during the turn. No, the characteristic quantity of the accelerated shot that implements the shape S is the 'quasi-deviation' of the acceleration waveform, but the variation of the other acceleration waveforms can be used (_ the average value of the acceleration can also be used. For each acceleration, In addition, the present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the spirit of the invention. Industrial applicability is as described above. The present invention is not only classified by visually observing the motion of the vehicle shown by the situation data, but can automatically and sinfully identify the vehicle motion indicated by the situation data. [Simplified illustration] 1 is a view schematically showing a vehicle behavior analysis system according to the present embodiment. FIG. 2 is a view showing components of the drive recorder according to the embodiment. FIG. 3 is a view showing a vehicle behavior analysis device according to the present embodiment. Functional configuration Fig. 4 is a flowchart showing 21 201219789f of the operation of the vehicle behavior analysis device according to the present embodiment. Fig. 5 is a view showing the vehicle in the embodiment. Diagram of the correspondence between the action and the feature quantity of each acceleration. [Description of main component symbols] 3: Detection unit 4: Reporting unit 5: Input unit 6: Communication unit (transmitter) 7: Loading and unloading recording unit 8: Information Processing unit 31: imaging unit 32: acceleration sensor 33: position sensor 41: illuminator (LED)
81 :中央處理單元/CPU 82 :内部記憶體/記憶體 83 : I/O緩衝電路 100 :行車記錄器 200 :車輛動作分析裝置 201 :狀況資料接收部/部分 202 :車輛動作鑑定部/部分 CN :連接器 D1 :狀況資料儲存部/部分 D2 :分析資料儲存部/部分 S1〜S5 :步驟 V :車輛/汽車 2281: central processing unit/CPU 82: internal memory/memory 83: I/O buffer circuit 100: driving recorder 200: vehicle motion analyzing device 201: status data receiving unit/portion 202: vehicle motion identifying unit/part CN : Connector D1 : Status data storage unit / Part D2 : Analysis data storage unit / Part S1 to S5 : Step V : Vehicle / Car 22