JP4173997B2

JP4173997B2 - Driving state determination processing method, driving state determination device, and driving state determination processing program

Info

Publication number: JP4173997B2
Application number: JP2003023219A
Authority: JP
Inventors: 悠萩原; 一哉鈴木
Original assignee: Fujitsu Ltd
Current assignee: Fujitsu Ltd
Priority date: 2003-01-31
Filing date: 2003-01-31
Publication date: 2008-10-29
Anticipated expiration: 2023-01-31
Also published as: JP2004234418A

Description

【０００１】
【発明の属する技術分野】
本発明は、車両の運転状態を判定する方法および装置に関する。さらに詳しくは、車両の走行情報のうち特に車両速度変化と変化時間の傾向とを利用して車両の運転状態を分析し、特定の運転状態を判定する方法および装置に関する。
【０００２】
【従来の技術】
従来、高速道路走行中の低刺激状態や居眠り運転状態等に代表される意識低下状態の検出は、車両のステアリングホイールの舵角センサや方位センサを用いて、車両方向の変化傾向について平常状態の傾向との差異を比較することにより行っていた。
【０００３】
例えば、居眠り運転状態では、意識状態の低下に伴い車両操縦機器類の操作が緩慢となり、かつ外部環境に対する注意力や意識力が低下するという特徴がある。この特徴を利用し、車両の操舵操縦について、意識低下状態では操作量の周波数スペクトルが低周波に遷移するという傾向が現れることから、平常状態を基準として比較した結果により、意識低下状態を検出していた。
【０００４】
また、車両運行状態を分析する従来手法の一つとして、車両の燃料消費量を計測したデータおよび速度データから燃料消費率を計算し、その値が基準値よりも高い場合には高燃費な運転が行われていると判断して記録し、運転者にその旨を通知する手法がある（特許文献１参照）。
【０００５】
【特許文献１】
特開平１０−６９５５５号公報
【０００６】
【発明が解決しようとする課題】
車両の特定の運転状態を判定する従来手法では、車両の操舵状況を把握するためにステアリングホイールの舵角センサや方位センサ等が必要となってしまい、汎用的な車載型情報処理機器だけでは車両運転状態の解析が行えないという問題があった。
【０００７】
また、車両運転状態におけるマクロ的な時間要因が考慮されないため、道路形状や交通状況による減速等の微少な速度変化により長時間かけて大きな速度変化を行った場合や、交通状況により急激な減速等を行った場合等の通常の運転状態を意識低下状態と分別できずに誤認識してしまうなどの問題もあった。
【０００８】
本発明は、これらの問題を解決するために、一定の条件により車両が巡航中と決定される速度帯域と意識低下運転時に速度の変動傾向が継続し得る時間範囲とを設定し、設定した速度帯域および時間範囲と車速とをもとに、特定の運転状態を簡易かつ高精度で判定することを目的とする。
【０００９】
【課題を解決するための手段】
本発明の解決課題と同様の課題を解決する手法として、すでに、運転状態を判定する手法を提案している（特願２００２−１２３４２９号）。この手法は、車速の任意時区間において移動平均による巡航速度帯域を定義し、しかる後に該当帯域に対して一定の傾向をもつ区間を抽出することにより特殊な運転状態を簡易かつ高精度で判定するものである。
【００１０】
しかし、既提案の処理手法による処理実行には、コンピュータなどの高度な情報処理機能を備えた機器が適している。例えば携帯型情報端末や車載端末のような小型の汎用情報処理機構のように処理能力や記憶容量に制限があるような機器においては、実行時の処理負荷が大きく解析処理の実行が難しい。車両走行中にリアルタイムで処理を実行しようとする場合には、無線通信手段等によって外部の情報処理装置に走行データをリアルタイムで送信し、外部処理する必要があった。このため、実際にリアルタイムで特定の運転状態を解析しようとすると、通信コスト、通信手段の用意等の点でコストがかさんでしまうという問題が生じることから、かかる処理におけるコスト低減が望まれている。
【００１１】
また、特定の運転状態のうち、特に意識低下状態の判定は、当該状態の発生時点で運転者の覚醒を促すことができる手法のほうが効果が高いことから、本来、意識低下状態の判定はリアルタイムに実現できることが望ましい。したがって、通信に拠らずに、車両上の廉価な処理機構において処理を直接実行できるような手法の実現が望まれている。
【００１２】
本発明は、前記の要求に鑑み、すでに提案した前記手法を発展させたものである。本発明は、意識低下状態では、所定の速度帯域および時間範囲における車両速度の変化傾向の特性を利用して、該当する状態を判定する。すなわち意識低下状態での運転時には運転者の加速器操作に特徴があり、この特徴が車両の車速に反映される特性を利用して当該状態を判定するものである。
【００１３】
具体的には、本発明は、一定の条件により決定される通常走行の速度帯域および意識低下運転時に速度の変動傾向が継続し得る時間範囲を設定し、設定した速度帯域および時間範囲と車速とをもとに、車速の移動平均値の変化傾向（加速度）に関する傾き（正負）が反転する点が出現する時間間隔および当該反転点の車速値の変化傾向を用いて意識低下状態という特定の運転状態の区間を判定することを特徴とする。
【００１４】
本発明は、車両の特定の運転状態を判定するため以下の処理を行う。
【００１５】
本発明は、車両の車速情報および前記車速情報の取得時間情報を取得し、前記車速情報をもとに予め定められた区間の移動平均値を算出し、前記移動平均値と当該移動平均値の直前の移動平均値との移動平均差分値を算出し、前記移動平均値および前記移動平均差分値を前記取得時間情報に対応付けて判定用データとして保持する。そして、前記判定用データのうち、前記移動平均差分値の符号が当該移動平均差分値の直前の移動平均差分値の符号と異なるものを抽出点として蓄積しておき、前記抽出点の移動平均差分値の符号が、正から負へ転換した抽出点、負から正へ転換した抽出点、および正から負へ転換した抽出点の並びで構成される区間を抽出し、前記抽出した区間の時間長が予め定められた意識低下状態である最小の区間を示す最小時間範囲内であり、かつ前記抽出した区間における最大移動平均値と最小移動平均値との差である区間内平均速度差が予め定められた異常規定用減速幅を示す異常判定用閾値以上であり、かつ前記抽出した区間における最低速度が予め定められた走行状態規定用閾値以上ある場合に、前記抽出した区間を意識低下状態を示す単一波形区間と判定する。
【００１６】
さらに、前記単一波形区間と当該単一波形区間より時系列上前にある単一波形区間との間隔が、予め定められた間隔判定用時間範囲内である場合に、前記時系列上前にある単一波形区間の始端から前記単一波形区間の終端までの区間を意識低下状態が連続する状態を示す連続波形区間と判定し、前記判定結果または前記判定結果を用いた情報を出力する。
【００１７】
さらに、本発明は、上記の処理を行い、前記連続波形区間を構成する単一波形区間群において、時系列上後にある前記単一波形区間の時間長および区間内平均速度差の両方の値が、時系列上前にある前記単一波形区間の時間長および区間内速度差より大きい場合に、前記単一波形区間群のうち、時系列上最前にある前記単一波形区間の始端から時系列上最後にある前記単一波形区間の終端までの区間を、意識低下状態が連続する状態を高精度に示す高精度連続波形区間と判定する。
【００１８】
本発明は、既提案手法（特願２００２−１２３４２９号）と同様に、対象車両の車速の変化傾向を測定して意識低下状態を判定するものであって、運転者の車両操作状態に関する情報を用いる従来の手法と異なるものである。
【００１９】
しかし、既提案手法が、設定した巡航速度帯域より車速が低下しているかどうかを速度差をもとに意識低下状態を判定するのに対し、本発明は、かかる数値比較処理を行わず、所定の通常走行速度帯域および時間範囲内で車速の加速と減速とが反転した時点の出現をもとに当該状態を判定する点が異なる。
【００２０】
これにより、本発明は、特殊な操舵操作センサを備えることなく汎用的な処理機構により特殊な運転状態を判定することが可能となる。
【００２１】
さらに、本発明は、車速およびその取得時間の情報のみを必要とすることから、デジタル式運行記録計等の汎用的な車載機において処理を実現することが可能であるとと共に、処理実行時に保持するデータ量および処理ステップ数が少ないという特徴をもつ。
【００２２】
したがって、車載型の汎用的な処理機構において、本発明の処理自体をリアルタイムに実行することが可能となり、意識低下状態を効果的に判定することが可能となる。運転者の実況に即したリアルタイムの運行指導を低コストに実現できる。
【００２３】
本発明の各手段または機能または要素をコンピュータにより実現するプログラムは、コンピュータが読み取り可能な、可搬媒体メモリ、半導体メモリ、ハードディスクなどの適当な記録媒体に格納することができ、これらの記録媒体に記録して提供され、または、通信インタフェースを介して種々の通信網を利用した送受信により提供されるものである。
【００２４】
【発明の実施の形態】
以下に、本発明に好適な実施の形態について図面にもとづいて説明する。
【００２５】
図１に、本発明を実現する処理装置の構成例を示す。運転状態判定装置１は、運行車両２の運行状況を解析し、特定の運転状態として意識低下状態を判定する処理装置であって、解析対象である運行車両２に搭載される。
【００２６】
運転状態判定装置１は、車速センサ１１、道路種別判定手段１２、車載記憶装置１３、走行状態判定手段１４、判定条件設定手段１５、および可搬性記憶媒体１６を備える。
【００２７】
また、運転状態判定装置１は、運行車両２に備えられたモニタ、スピーカ等の出力装置（図示しない）に接続され、判定結果１７など任意の情報を音声または画像等により出力することができる。
【００２８】
車速センサ１１は、所定の時間ごとに運行車両２の運行速度（車速）を測定し、車速値Ｖz およびその測定時刻を含む車速データを取得し、車載記憶装置１３に記憶する手段である。
【００２９】
道路種別判定手段１２は、運行車両２の車速データから運行車両２が走行中の道路種別を判定する手段である。道路種別判定手段１２は、例えば、一般道路（市街道、国道）、高速道路（一般高速道、首都圏高速道）等の道路種別ごとに、通常の巡航中の平均車速等の統計的情報をもとに定めた道路種別規定用速度閾値Ｖthを記憶して、道路種別規定用速度閾値Ｖthをもとに走行中の道路種別を判定する。
【００３０】
なお、道路種別の判定基準は、道路種別規定用速度閾値Ｖthではなく、道路種別を決定するスイッチ情報等の時間情報であってもよい。
【００３１】
車載記憶装置１３は、車速センサ１１の測定したデータ（車速データ）、道路種別判定手段１２の判定結果、および走行状態判定手段１４の判定結果を記憶する手段である。また、車載記憶装置１３は、所定の契機または間隔で、記憶している種々の情報を可搬性記憶媒体１６に格納する。
【００３２】
走行状態判定手段１４は、車載記憶装置１３を参照して運行車両２の走行状態を解析し、所定の判定条件をもとに運行車両２が特定の運転状態の一つである意識低下状態にあるか否かを判定する手段であって、移動平均算出部１４１および意識低下状態判定部１４２を備える。
【００３３】
移動平均算出部１４１は、車載記憶装置１３に記憶された運行車両２の車速値Ｖz から、車速の移動平均値（平均速度）Ｖa および移動平均値Ｖa の差分値（加速度）Ｖb を算出する手段である。
【００３４】
意識低下状態判定部１４２は、移動平均算出部１４１の算出結果をもとに、判定条件を参照して運行車両２の運行状態から意識低下状態を抽出する手段である。意識低下状態判定部１４２は、判定条件として、走行状態規定用閾値Ｖlm、巡航状態規定用閾値Ｖath 、異常規定用減速幅Ｖthl 、意識低下規定用最大時間幅Ｔt 、意識低下規定用最小時間幅Ｔb 、波形連続規定用区間幅Ｔd 、および移動累積集計区間設定値ｎath を規定しておく。
【００３５】
走行状態規定用閾値Ｖlmは、運行車両２が巡航中であるか否かを判定するための値である。巡航状態規定用閾値Ｖath は、運行車両２の減速が通常の巡航範囲内であるか否かを判定するための値である。
【００３６】
異常規定用減速幅Ｖthl は、巡航中の運行車両２の減速が通常の巡航中の減速範囲であるか否かを判定するための値である。意識低下規定用最大時間幅Ｔt および意識低下規定用最小時間幅Ｔb は、抽出された判定対象となる区間の時間範囲が意識低下状態であるか否かを判定するための最大範囲および最小範囲を規定する値である。
【００３７】
波形連続規定用区間幅Ｔd は、意識低下状態と判定した区間が連続するか否かを判定するための値である。移動累積集計区間設定値ｎath は、移動平均値の累積集計を行う区間を設定するための値である。
【００３８】
判定条件設定手段１５は、走行状態判定手段１４で用いられる判定条件を設定する手段である。判定条件設定手段１５は、運転状態判定装置１に接続する入力装置（図示しない）もしくは通信装置（図示しない）、可搬性記憶媒体１６等を介して入力した情報を用いて判定条件を設定する。
【００３９】
可搬性記憶媒体１６は、車載記憶装置１３に記憶される種々の情報を記憶する読み取りまたは書き込み可能な媒体であって、例えばＩＣカードで実現する。
【００４０】
以下、運転状態判定装置１の処理動作について図２ないし図６を用いて説明する。ここで、運転状態判定装置１は、道路種別として一般道路および高速道路を規定するものとし、高速道路であるか否かの道路種別規定用速度閾値Ｖthを規定して、運行車両２の車速値Ｖz が道路種別規定用速度閾値Ｖthを超えた場合に道路種別＝高速道路と判定する。
【００４１】
図２に、本発明の処理概要を示す。
【００４２】
移動平均算出部１４１により、車速センサ１１で計測され車載記憶装置１３に記憶された運行車両２の車速データのうち車速値Ｖz から移動平均値Ｖa 、差分値Ｖb 等の各種変数を算出する（ステップＳ２０）。
【００４３】
そして、意識低下状態判定部１４２により、判定条件を参照し、算出された変数（移動平均値Ｖa 、差分値Ｖb ）をもとに意識低下状態であるか否かを判定する（ステップＳ２１〜Ｓ２４）。
【００４４】
すなわち、意識低下状態判定部１４２では、差分値Ｖb の符号の正負が転換する点（極点）を抽出し、当該点が３点になるまで蓄積する（ステップＳ２１）。その後、蓄積した３点について、意識低下状態の最小区間を規定する条件に該当するか否かを判定する（ステップＳ２２）。
【００４５】
さらに、３点が条件に該当する区間（該当区間）が、当該区間より時間的に前にある該当区間と所定時間範囲内で連続しているか否かを判定する（ステップＳ２３）。さらに、これらの連続する条件該当区間（連続区間）が、高精度で連続する意識低下状態を規定する条件に該当するか否かを判定する（ステップＳ２４）。
【００４６】
そして、当該３点が最小区間を規定する条件に該当しない場合には（ステップＳ２２）、当該３点で形成される区間は通常走行であると判定する（ステップＳ２５）。
【００４７】
一方、当該３点が前記条件に該当するが（ステップＳ２２）、それ以前にある該当区間とは連続しない場合には（ステップＳ２３）、現該当区間を単一波形区間（第１の区間）と判定し、判定結果を車載記憶装置１３に記録する（ステップＳ２６）。
【００４８】
さらに、現該当区間が、それ以前にある該当区間と連続するが（ステップＳ２３）、連続する該当区間が連続する意識低下状態の高精度の条件に該当しない場合には（ステップＳ２４）、現該当区間と連続するそれ以前の該当区間とで形成される区間を連続波形区間（第２の区間）と判定し、判定結果を車載記憶装置１３に記録する（ステップＳ２７）。
【００４９】
一方、現該当区間が、それ以前にある該当区間と連続し（ステップＳ２３）、かつ前記連続する該当区間群が連続する意識低下状態の高精度の条件に該当する場合には（ステップＳ２４）、前記条件を満たす該当区間群で形成される区間を高精度連続波形区間（第３の区間）と判定し、判定結果を車載記憶装置１３に記録する（ステップＳ２８）。
【００５０】
図３に、前記ステップＳ２０の移動平均算出処理について、より詳細な処理フローを示す。
【００５１】
道路種別判定手段１２は、車速センサ１１から逐次計測された車速値Ｖz （ｚ＝１，２，３…）および各車速値Ｖz の時間情報を取得し（ステップＳ３０）、道路種別規定用速度閾値Ｖthを基準に当該車速値Ｖz の大小判定を行う（ステップＳ３１）。車速値Ｖz が道路種別規定用速度閾値Ｖthを超えた場合に、ステップＳ３２以降の処理が行われる。
【００５２】
移動平均算出部１４１は、現時点において、取得している車速値Ｖz のデータ数ｎa が所定の移動累積集計区間設定値ｎath に達しているか否かを判定する（ステップＳ３２、Ｓ３４、Ｓ３８）。
【００５３】
まず、データ数ｎa が移動累積集計区間設定値ｎath −１より小さければ（ステップＳ３２）、取得した車速値Ｖz を累積する（ステップＳ３３）。
【００５４】
また、データ数ｎa が移動累積集計区間設定値ｎath −１と等しければ（ステップＳ３４）、取得した車速値Ｖz を累積し（ステップＳ３５）、その後に現時点までに累積された車速値Ｖz の移動平均値Ｖa を算出し（ステップＳ３６）、続いて、当該車速値Ｖz の直前に算出された移動平均値Ｖa-1 を用いて、差分値Ｖb （Ｖb ＝Ｖa −Ｖa-1 ）を算出する（ステップＳ３７）。
【００５５】
また、データ数ｎa が移動累積集計区間設定値ｎath と等しければ（ステップＳ３８）、現時点で累積しているデータのうち最前のものを破棄し（ステップＳ３９）、その後に、ステップＳ３５からステップＳ３７までの処理を行う。
【００５６】
運行車両２の走行が終了するまで（ステップＳ３１０）、車速センサ１１から車速値Ｖz が入力されるたびにステップＳ３０からステップＳ３９までの処理を繰り返し実行する。また、次に説明する意識低下状態判定処理においても、常に最新の移動平均値Ｖa 、差分値Ｖb を用いて処理を行うものとする。
【００５７】
図４および図５に、前記ステップＳ２１〜Ｓ２８の意識低下状態判定処理について、より詳細な処理フローを示す。
【００５８】
意識低下状態判定部１４２では、移動平均算出部１４１から、移動平均値Ｖa および差分値Ｖb を取得する（ステップＳ４０）。この際、取得した差分値Ｖb と直前の差分値Ｖb-1 とで符号の正負が反転しているかどうかを調べ（ステップＳ４１）、差分値Ｖb の符号が直前の差分値Ｖb-1 の符号と等しければ、取得した移動平均値Ｖa および差分値Ｖb については以降の判定処理を行わず、ステップＳ４０の処理へ戻る。
【００５９】
一方、差分値Ｖb の符号が直前の差分値Ｖb-1 の符号と異なっていれば、取得した当該時点での差分値Ｖb およびその移動平均値Ｖa ならびにその取得時間とを意識低下状態判定用データ（抽出点）としてさらに抽出する（ステップＳ４２）。
【００６０】
続いて、現状で抽出されている抽出点数を判定し（ステップＳ４３）、抽出点数＝３でなければ、以下の判定処理は行わず、ステップＳ４０の処理へ戻る。一方、抽出点数が３点であれば、３つの抽出点の移動平均値Ｖa （ここで、時系列順にＶa-2 、Ｖa-1 、Ｖa とする）に対して、意識低下状態を規定する判定条件を用いて判定を行う。
【００６１】
まず、３つの移動平均値Ｖa-2 、Ｖa-1 、Ｖa が、条件「Ｖa ＞Ｖa-1 かつＶa-2 ＞Ｖa-1 」を満たすか否かの判定処理（ステップＳ４４）、および差分Ｖa −Ｖa-2 が移動平均値変化における巡航状態を規定する巡航状態規定用閾値Ｖath を満たすか否かの判定を行う（ステップＳ４５）。
【００６２】
そして、ステップＳ４４およびＳ４５の条件に該当する場合は、移動平均値Ｖa ないしＶa-2 の最高移動平均値（最高速度）から最低移動平均値（最低速度）の差分（速度差）Ｖl を算出し（ステップＳ４６）、さらに、差分Ｖl が速度変化（減速）の異常状態を規定する異常規定用減速幅Ｖthl を上回るか否かの判定（ステップＳ４７）、および最低移動平均値（最低速度）が車両の通常走行を規定する走行状態規定用閾値Ｖlmを上回るか否かの判定を行う（ステップＳ４８）。
【００６３】
ステップＳ４７およびＳ４８の条件に該当する場合は、３つの抽出点で形成される区間（抽出区間）の時間情報から、当該３抽出点の時間範囲Ｔq を算出し（ステップＳ４９）、時間範囲Ｔq が意識低下状態の時間範囲を規定する意識低下規定用最大時間幅Ｔt および意識低下規定用最小時間幅Ｔb に対し、Ｔt ＞Ｔq ＞Ｔb を満たすか否かを判定する（ステップＳ４１０）。
【００６４】
前記のステップＳ４４からステップＳ４１０までの判定処理において、いずれかの条件に該当しなかった場合は、抽出している中で時系列上最も古い抽出点のデータ（Ｖa-2 ）を破棄し（ステップＳ４１１）、以降の処理を行わず、ステップＳ４０の処理へ戻る。
【００６５】
一方、ステップＳ４４からステップＳ４１０までの判定処理において、全ての条件に該当した場合は、当該３抽出点の区間が意識低下状態を規定する単一波形区間（第１の区間）であると判定し、判定結果と検出時間とを関連付けて車載記憶装置１３へ記録し、音声、画像等の出力装置より出力する（ステップＳ４１２）。判定結果は、例えば、音声データとして出力されて運行車両２の運転者へ通知され、または、画像データとして出力されて運行車両２に備えられた表示装置へ表示される。
【００６６】
ステップＳ４１２の処理完了後に、抽出点のうち時系列的に最も古い抽出点（Ｖa-2 ）を破棄し（ステップＳ４１３）、意識低下状態と判定された現抽出区間が、同一運行上第２番目以降の区間であるか否かの判定を行う（ステップＳ４１４）。現抽出区間が第１番目の区間であれば、以降の処理を行わずに、ステップＳ４０の処理へ戻る。
【００６７】
一方、現抽出区間が第２番目以降の区間であれば、現抽出区間より時系列上前にある抽出区間（第１の区間）の終端である抽出点（Ｖa ）と現抽出区間の始端である抽出点（Ｖa-2 ）との時間差Ｔs を算出し（ステップＳ４１５）、時間差Ｔs が、単一波形区間が連続している状態（連続波形区間）を規定する波形連続規定用区間幅Ｔd 内であるか否かの判定を行う（ステップＳ４１６）。そして、Ｔs ＜Ｔd であれば、時系列上前の抽出区間の始端（抽出点）から現抽出区間の終端（抽出点）までの区間を意識低下運転状態が連続する連続波形区間（第２の区間）と判定し、判定結果と検出時間とを関連付けて車載記憶装置１３へ記録し、音声や画像等の出力装置より出力する（ステップＳ４１７）。一方、Ｔs ＜Ｔd でなければ、第２の区間と判定せず、また以下の処理を行わずに、ステップＳ４０の処理へ戻る。
【００６８】
続いて、第２の区間と判定された抽出区間群について、時系列上前にある（古い）抽出区間の区間時間長Ｔqtと時系列上後にある（新しい）抽出区間の区間時間長Ｔqu、および、時系列上前の抽出区間の区間速度差Ｖltと時系列上後の抽出区間の区間速度差Ｖluについて、条件「Ｔqt＜ＴquかつＶlt＜Ｖlu」を満たすか否かを判定する（ステップＳ４１８）。
【００６９】
そして、条件に該当する場合は、該当する抽出区間群のうち時系列的に最前の該当区間の始端（抽出点）から最後の該当区間の終端（抽出点）までの区間を、意識低下状態が連続する状態が高精度に判定できる高精度連続波形区間（第３の区間）と判定し、判定結果と検出時間とを関連付けて車載記憶装置１３へ記録し、音声、画像等の出力装置より出力する（ステップＳ４１９）。一方、条件に該当しない場合は、第３の区間を判定せず、ステップＳ４０の処理へ戻る。
【００７０】
以上の処理は、移動平均値Ｖa 、差分値Ｖb が取得されるたびに、運行車両２の走行終了まで繰り返し実行する（ステップＳ４２０）。
【００７１】
図６に、本発明において判定される運行車両２の走行状態例を示す。図６において、縦軸は移動平均値Ｖa （km/h）を、横軸は運行経過の時間情報（hh：mm）を示す。
【００７２】
図４のステップＳ４１の判定条件に該当する抽出点の移動平均値Ｖa について、走行状態規定用閾値Ｖlmを常に上回る３つの連続し、かつ移動平均値Ｖa の符号転換点である抽出点の移動平均値Ｖa （Ｖa 、Ｖa-1 、Ｖa-2 ）を判定対象とし、これらの移動平均値Ｖa （Ｖa 、Ｖa-1 、Ｖa-2 ）が全て一定の異常規定用減速幅Ｖthl 以上であり、かつ、その時間範囲Ｔq が意識低下状態規定用最大時間幅Ｔt および意識低下状態規定用最小時間幅Ｔb の範囲内である区間（波形）を意識低下状態の単一波形区間と分別し抽出していることがわかる。
【００７３】
なお、走行状態判定手段１４は、判定された単一波形区間（第１の区間）、連続波形区間（第２の区間）、または高精度連続波形区間（第３の区間）を、常に全ての区間について判定ごとに記録または出力する必要はない。例えば、意識低下状態を最も高精度に判定した第３の区間のみを記録および出力するなど、任意の区間を検出した場合のみ記録または出力してもよい。
【００７４】
さらに、判定の結果、記録された３つの区間の出力形態は、音声等によるリアルタイムの警告以外に、記録された判定結果情報を用いて事後に出力するものであってもよい。この場合に、車載記憶装置１３に記憶された判定結果は、可搬性記憶媒体１６にも記録されるものとする。
【００７５】
図７に、判定結果を事後出力する場合の出力例について示す。図７に示す判定結果の出力例において、第１の区間（Ｐ１）は、意識低下状態の判定条件を満たす波形に該当する単一区間の開始から終了までの時間範囲が記録される。第２の区間（Ｐ２）および第３の区間（Ｐ３）は、連続して発生した場合の最前の第１の区間の開始より最終の第１の区間の終了までの時間範囲がそれぞれ記録される。このように、運転状態判定装置１の判定結果すなわち意識低下状態の各区間の記録は、例えばデジタル式運行記録計等の走行車速情報と関連付けて出力することができる。
【００７６】
以上、本発明をその実施の態様により説明したが、本発明はその主旨の範囲において種々の変形が可能である。
【００７７】
例えば、本発明の実施の形態において、運転状態判定装置１が車速センサ１１を備えるものとして説明した。しかし、運転状態判定装置１を搭載する運行車両２に車速が測定可能なセンサが備えられている場合には、運転状態判定装置１が車速センサ１１を備える必要はない。
【００７８】
また、運転状態判定装置１は、通信処理手段に接続可能として、車載記憶装置１３に記憶される情報を送信し、または、判定条件設定手段１５で用いる情報を受信できるようにしてもよい。
【００７９】
また、運転状態判定装置１の走行状態判定手段１４は、運行車両２の外部処理機構として構成され、可搬性記憶媒体１６の情報を用いて事後に判定処理を実行するものであってもよい。
【００８０】
本発明の形態および実施例の特徴を列記すると以下のとおりである。
【００８１】
（付記１）車両の特定の運転状態を判定する処理方法であって、
車両の車速情報および前記車速の取得時間情報を取得し、
前記車速情報をもとに任意区間の移動平均値を算出し、前記移動平均値と当該移動平均値の直前の移動平均値との移動平均差分値を算出し、
前記移動平均値および前記移動平均差分値を前記取得時間情報に対応付けて判定用データとして保持し、
前記判定用データのうち、前記移動平均差分値の符号が当該移動平均差分値の直前の移動平均差分値の符号と異なるものを抽出点として蓄積し、
前記抽出点の移動平均差分値の符号が、正から負へ転換した抽出点、負から正へ転換した抽出点、および正から負へ転換した抽出点の並びで構成される区間について、前記抽出区間の時間長が任意の時間範囲内であり、かつ前記区間における最大移動平均値と最小移動平均値との差である区間内平均速度差が任意の減速範囲内であり、かつ前記移動平均値差分値の符号が正から負へ転換した２つの前記抽出点の移動平均値の差分が任意の速度範囲内である場合に、前記区間を意識低下状態を示す単一波形区間と判定し、
前記判定結果または前記判定に用いた情報を出力する
ことを特徴とする運転状態判定処理方法。
【００８２】
（付記２）前記付記１に記載の運転状態判定処理方法であって、
前記単一波形区間が、前記単一波形区間より時系列上前にある単一波形区間と任意の時間範囲内に発生している場合に、前記時系列上前にある単一波形区間の始端から前記単一波形区間の終端までの区間を意識低下状態が連続する状態を示す連続波形区間と判定し、
前記判定結果または前記判定に用いた情報を出力する
ことを特徴とする運転状態判定処理方法。
【００８３】
（付記３）前記付記２に記載の運転状態判定処理方法であって、
前記連続波形区間を構成する単一波形区間群において、時系列上後にある前記単一波形区間の時間長および区間内平均速度差の両方の値が、時系列上前にある前記単一波形区間の時間長および区間内速度差より大きい場合に、前記単一波形区間群のうち、時系列上最前にある前記単一波形区間の始端から時系列上最後にある前記単一波形区間の終端までの区間を意識低下状態が連続する高精度の状態を示す高精度連続波形区間と判定し、
前記判定結果または前記判定に用いた情報を出力する
ことを特徴とする運転状態判定処理方法。
【００８４】
（付記４）車両の特定の運転状態を判定する処理装置であって、
車両の車速情報および前記車速の取得時間情報を取得する車速情報取得手段と、
前記車速情報をもとに任意区間の移動平均値を算出し、前記移動平均値と当該移動平均値の直前の移動平均値との移動平均差分値を算出する移動平均算出手段と、
前記移動平均値および前記移動平均差分値を前記取得時間情報に対応付けて判定用データとして保持する記憶手段と、
前記判定用データのうち、前記移動平均差分値の符号が当該移動平均差分値の直前の移動平均差分値の符号と異なるものを抽出点として蓄積し、前記抽出点の移動平均差分値の符号が、正から負へ転換した抽出点、負から正へ転換した抽出点、および正から負へ転換した抽出点の並びで構成される区間について、前記抽出区間の時間長が任意の時間範囲内であり、かつ前記区間における最大移動平均値と最小移動平均値との差である区間内平均速度差が任意の減速範囲内であり、かつ前記移動平均値差分値の符号が正から負へ転換した２つの前記抽出点の移動平均値の差分が任意の速度範囲内である場合に、前記区間を意識低下状態を示す単一波形区間と判定し、前記判定結果または前記判定に用いた情報を出力する意識低下状態判定手段とを備える
ことを特徴とする運転状態判定装置。
【００８５】
（付記５）車両の特定の運転状態を判定する処理を、コンピュータに実行させるためのプログラムであって、
車両の車速情報および前記車速の取得時間情報を取得する処理と、
前記車速情報をもとに任意区間の移動平均値を算出し、前記移動平均値と当該移動平均値の直前の移動平均値との移動平均差分値を算出する処理と、
前記移動平均値および前記移動平均差分値を前記取得時間情報に対応付けて判定用データとして保持する処理と、
前記判定用データのうち、前記移動平均差分値の符号が当該移動平均差分値の直前の移動平均差分値の符号と異なるものを抽出点として蓄積する処理と、
前記抽出点の移動平均差分値の符号が、正から負へ転換した抽出点、負から正へ転換した抽出点、および正から負へ転換した抽出点の並びで構成される区間について、前記抽出区間の時間長が任意の時間範囲内であり、かつ前記区間における最大移動平均値と最小移動平均値との差である区間内平均速度差が任意の減速範囲内であり、かつ前記移動平均値差分値の符号が正から負へ転換した２つの前記抽出点の移動平均値の差分が任意の速度範囲内である場合に、前記区間を意識低下状態を示す単一波形区間と判定する処理と、
前記判定結果または前記判定に用いた情報を出力する処理とを、
コンピュータに実行させる
ことを特徴とする運転状態判定処理プログラム。
【００８６】
（付記６）車両の特定の運転状態を判定する処理を、コンピュータに実行させるためのプログラムを記録した記録媒体であって、
車両の車速情報および前記車速の取得時間情報を取得する処理と、
前記車速情報をもとに任意区間の移動平均値を算出し、前記移動平均値と当該移動平均値の直前の移動平均値との移動平均差分値を算出する処理と、
前記移動平均値および前記移動平均差分値を前記取得時間情報に対応付けて判定用データとして保持する処理と、
前記判定用データのうち、前記移動平均差分値の符号が当該移動平均差分値の直前の移動平均差分値の符号と異なるものを抽出点として蓄積する処理と、
前記抽出点の移動平均差分値の符号が、正から負へ転換した抽出点、負から正へ転換した抽出点、および正から負へ転換した抽出点の並びで構成される区間について、前記抽出区間の時間長が任意の時間範囲内であり、かつ前記区間における最大移動平均値と最小移動平均値との差である区間内平均速度差が任意の減速範囲内であり、かつ前記移動平均値差分値の符号が正から負へ転換した２つの前記抽出点の移動平均値の差分が任意の速度範囲内である場合に、前記区間を意識低下状態を示す単一波形区間と判定する処理と、
前記判定結果または前記判定に用いた情報を出力する処理とを、
コンピュータに実行させる処理プログラムを記録する運転状態判定処理プログラム記録媒体。
【００８７】
【発明の効果】
以上説明したように、本発明によれば、低負荷もしくは低記憶容量の環境下においても意識低下状態を検出することが可能となり、汎用的な車載型情報処理機構や携帯式情報端末等により意識低下状態判定処理を実現することができる。
【００８８】
また、本発明によれば、車両上で即時に意識低下状態を判定する手法を、より低コストに実現することが可能になり、運行管理の経済性が向上する。
【図面の簡単な説明】
【図１】本発明を実現する処理装置の構成例を示図である。
【図２】本発明の処理概要を示す図である。
【図３】移動平均算出処理の詳細な処理フローを示す図である。
【図４】意識低下状態判定処理の詳細な処理フローを示す図である。
【図５】意識低下状態判定処理の詳細な処理フローを示す図である。
【図６】本発明において判定される運行車両の走行状態例を示す図である。
【図７】判定結果を事後出力する場合の出力例を示す図である。
【符号の説明】
１運転状態判定装置（車載機）
１１車速センサ
１２道路種別判定手段
１３車載記憶装置
１４走行状態判定手段
１４１移動平均算出部
１４２意識低下状態判定部
１５判定条件設定手段
１６可搬性記憶媒体
１７判定結果（音声、画像等）
２運行車両[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a method and an apparatus for determining a driving state of a vehicle. More specifically, the present invention relates to a method and an apparatus for analyzing a driving state of a vehicle by using a vehicle speed change and a tendency of change time in the running information of the vehicle to determine a specific driving state.
[0002]
[Prior art]
Conventionally, detection of a low consciousness state represented by a low stimulus state or a drowsiness driving state on an expressway is performed by using a steering angle sensor or a direction sensor of a steering wheel of the vehicle, It was done by comparing the difference with the trend.
[0003]
For example, in the drowsy driving state, there is a feature that the operation of the vehicle control devices becomes slow as the consciousness state decreases, and the attention and consciousness to the external environment decrease. Using this feature, the steering spectrum of the vehicle tends to show that the frequency spectrum of the manipulated variable transitions to a low frequency in the low consciousness state, so the low consciousness state is detected based on the comparison result based on the normal state. It was.
[0004]
In addition, as one of the conventional methods for analyzing the vehicle operating status, the fuel consumption rate is calculated from the data that measures the fuel consumption of the vehicle and the speed data. There is a method of determining that the vehicle is being recorded and recording it, and notifying the driver accordingly (see Patent Document 1).
[0005]
[Patent Document 1]
JP-A-10-69555
[0006]
[Problems to be solved by the invention]
The conventional method for determining a specific driving state of a vehicle requires a steering wheel angle sensor, a direction sensor, and the like in order to grasp the steering state of the vehicle. There was a problem that analysis of the driving state could not be performed.
[0007]
In addition, since macro time factors in the driving state of the vehicle are not taken into account, when a large speed change is made over a long period of time due to a slight speed change such as deceleration due to road shape or traffic conditions, sudden deceleration etc. depending on traffic conditions There is also a problem that a normal driving state such as when performing a vehicle is misidentified without being distinguished from a state of reduced consciousness.
[0008]
In order to solve these problems, the present invention sets a speed range in which the vehicle is determined to be cruising under certain conditions and a time range in which the fluctuation tendency of the speed can continue during reduced consciousness driving. An object is to easily and accurately determine a specific driving state based on a band, a time range, and a vehicle speed.
[0009]
[Means for Solving the Problems]
As a technique for solving the same problem as the problem to be solved by the present invention, a technique for determining an operating state has already been proposed (Japanese Patent Application No. 2002-123429). This method defines a cruising speed band based on a moving average in an arbitrary section of vehicle speed, and then extracts a section having a certain tendency with respect to the corresponding band, thereby easily and accurately determining a special driving state. Is.
[0010]
However, a device having an advanced information processing function such as a computer is suitable for processing execution by the proposed processing method. For example, in a device with limited processing capability and storage capacity such as a small general-purpose information processing mechanism such as a portable information terminal or a vehicle-mounted terminal, the processing load at the time of execution is large and it is difficult to execute the analysis process. When processing is to be performed in real time while the vehicle is traveling, it is necessary to transmit the traveling data to an external information processing apparatus in real time by wireless communication means or the like and perform external processing. For this reason, when actually trying to analyze a specific operating state in real time, there arises a problem that the cost is increased in terms of communication cost, preparation of communication means, etc., and thus cost reduction in such processing is desired. Yes.
[0011]
In particular, the determination of the state of reduced consciousness is particularly effective in determining the state of reduced consciousness because the method that can prompt the driver to wake up at the time of occurrence of the state is more effective. It is desirable that it can be realized. Therefore, it is desired to realize a technique that can directly execute processing in an inexpensive processing mechanism on a vehicle without relying on communication.
[0012]
The present invention has been developed from the previously proposed method in view of the above requirements. According to the present invention, in a state of reduced consciousness, a corresponding state is determined using the characteristics of a change tendency of vehicle speed in a predetermined speed band and time range. That is, when driving in a state of reduced consciousness, there is a feature in the driver's accelerator operation, and this feature is determined using characteristics reflected in the vehicle speed of the vehicle.
[0013]
Specifically, the present invention sets a normal driving speed band determined by certain conditions and a time range in which the tendency of fluctuation in speed during consciousness reduction driving can continue, and the set speed band and time range and vehicle speed Based on the time interval at which a point where the slope (positive / negative) with respect to the change tendency (acceleration) of the moving average value of the vehicle speed reverses and the change tendency of the vehicle speed value at the inversion point is used as a specific driving state of reduced consciousness A state interval is determined.
[0014]
The present invention performs the following processing to determine a specific driving state of the vehicle.
[0015]
  The present invention provides vehicle speed information of the vehicle and the vehicle speed.informationAcquisition time information, and based on the vehicle speed informationPredeterminedThe moving average value of the section is calculated, the moving average difference value between the moving average value and the moving average value immediately before the moving average value is calculated, and the moving average value and the moving average difference value are used as the acquisition time information. Correspondingly stored as determination data. Then, among the determination data, the moving average difference value of the extraction point is accumulated as an extraction point in which the sign of the moving average difference value is different from the sign of the moving average difference value immediately before the moving average difference value. An interval in which the sign of the value consists of an extraction point converted from positive to negative, an extraction point converted from negative to positive, and an extraction point converted from positive to negativeExtractThe extractiondidThe length of the section isThe minimum indicating the minimum interval that is in a predetermined state of consciousness declineWithin the time range and saidExtractedThe average speed difference in the section, which is the difference between the maximum moving average value and the minimum moving average value in the section, isMore than a threshold for abnormality determination indicating a predetermined abnormality regulation deceleration widthAndThe minimum speed in the extracted section is equal to or greater than a predetermined threshold for defining driving conditionsIf there isExtractedJudge the section as a single waveform section indicating a state of reduced consciousness.
[0016]
  Further, the single waveform sectionAnd the relevantA single waveform section that is earlier in time series than a single waveform sectionThe interval is determined for a predetermined interval.Within time rangeIsIn this case, the interval from the beginning of the single waveform interval preceding the time series to the end of the single waveform interval is determined as a continuous waveform interval indicating a state in which the state of consciousness continues.And output the determination result or information using the determination result.To do.
[0017]
  Furthermore, the present invention performs the above-described processing, and in the single waveform section group constituting the continuous waveform section, the values of both the time length of the single waveform section and the average speed difference in the section after the time series are When the time length of the single waveform section preceding the time series and the speed difference within the section are larger, the time series from the beginning of the single waveform section preceding the time series in the single waveform section group. The section from the top to the end of the single waveform section,The state of consciousness continuesThe state with high accuracyIt is determined to be the high-precision continuous waveform section shown.
[0018]
The present invention, like the already proposed method (Japanese Patent Application No. 2002-123429), measures the change tendency of the vehicle speed of the target vehicle to determine the state of consciousness reduction. This is different from the conventional method used.
[0019]
However, while the previously proposed method determines a state of reduced consciousness based on the speed difference whether the vehicle speed is lower than the set cruise speed band, the present invention does not perform such numerical comparison processing, The difference is that the state is determined on the basis of the appearance of the time when the acceleration and deceleration of the vehicle speed are reversed within the normal travel speed band and the time range.
[0020]
Thereby, the present invention can determine a special driving state by a general-purpose processing mechanism without providing a special steering operation sensor.
[0021]
Furthermore, since the present invention requires only information on the vehicle speed and its acquisition time, the processing can be realized in a general-purpose in-vehicle device such as a digital operation recorder and held at the time of processing execution. The amount of data to be processed and the number of processing steps are small.
[0022]
Therefore, in the on-vehicle general-purpose processing mechanism, the processing itself of the present invention can be executed in real time, and the state of reduced consciousness can be determined effectively. Real-time operation guidance in line with the driver's actual situation can be realized at low cost.
[0023]
A program for realizing each means, function, or element of the present invention by a computer can be stored in an appropriate recording medium such as a portable medium memory, a semiconductor memory, or a hard disk that can be read by the computer. It is provided by recording, or is provided by transmission / reception using various communication networks via a communication interface.
[0024]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings.
[0025]
FIG. 1 shows a configuration example of a processing apparatus for realizing the present invention. The driving state determination device 1 is a processing device that analyzes the driving situation of the operating vehicle 2 and determines a state of reduced consciousness as a specific driving state, and is mounted on the operating vehicle 2 that is an analysis target.
[0026]
The driving state determination device 1 includes a vehicle speed sensor 11, a road type determination unit 12, an in-vehicle storage device 13, a traveling state determination unit 14, a determination condition setting unit 15, and a portable storage medium 16.
[0027]
In addition, the driving state determination device 1 is connected to an output device (not shown) such as a monitor and a speaker provided in the traveling vehicle 2 and can output arbitrary information such as the determination result 17 by voice or image.
[0028]
The vehicle speed sensor 11 is means for measuring the operation speed (vehicle speed) of the operating vehicle 2 at predetermined time intervals, obtaining vehicle speed data including the vehicle speed value Vz and the measurement time, and storing it in the in-vehicle storage device 13.
[0029]
The road type determination means 12 is a means for determining the road type on which the operating vehicle 2 is traveling from the vehicle speed data of the operating vehicle 2. For example, the road type determination means 12 obtains statistical information such as average vehicle speed during normal cruising for each road type such as general roads (city roads, national roads) and highways (general highways, metropolitan expressways). The road type prescribing speed threshold value Vth that was originally defined is stored, and the road type that is running is determined based on the road type prescribing speed threshold value Vth.
[0030]
The road type determination criterion may be time information such as switch information for determining the road type instead of the road type defining speed threshold Vth.
[0031]
The in-vehicle storage device 13 is a unit that stores data (vehicle speed data) measured by the vehicle speed sensor 11, a determination result of the road type determination unit 12, and a determination result of the traveling state determination unit 14. The in-vehicle storage device 13 stores various information stored in the portable storage medium 16 at a predetermined trigger or interval.
[0032]
The traveling state determination unit 14 analyzes the traveling state of the operating vehicle 2 with reference to the in-vehicle storage device 13, and enters the reduced consciousness state in which the operating vehicle 2 is one of specific driving states based on a predetermined determination condition. A means for determining whether or not there is a moving average calculation unit 141 and a reduced consciousness state determination unit 142.
[0033]
The moving average calculating unit 141 calculates a moving average value (average speed) Va of vehicle speed and a difference value (acceleration) Vb of the moving average value Va from the vehicle speed value Vz of the operating vehicle 2 stored in the in-vehicle storage device 13. It is.
[0034]
The reduced consciousness state determination unit 142 is a unit that extracts a reduced consciousness state from the operation state of the traveling vehicle 2 with reference to the determination condition based on the calculation result of the moving average calculation unit 141. The consciousness lowering state determination unit 142 includes, as determination conditions, a driving state prescribing threshold Vlm, a cruising state prescribing threshold Vath, an abnormality prescribing deceleration width Vthl, a consciousness lowering prescription maximum time width Tt, and a consciousness lowering prescription minimum time width Tb. In addition, a waveform continuation defining interval width Td and a movement accumulation total interval setting value nath are defined.
[0035]
The threshold value Vlm for defining the traveling state is a value for determining whether or not the operating vehicle 2 is cruising. The threshold Vath for cruise state regulation is a value for determining whether or not the deceleration of the operating vehicle 2 is within the normal cruise range.
[0036]
The abnormality regulating deceleration width Vthl is a value for determining whether or not the deceleration of the traveling vehicle 2 during cruising is within the deceleration range during normal cruising. The maximum time width Tt for lowering consciousness and the minimum time width Tb for lowering consciousness specify the maximum range and the minimum range for determining whether or not the extracted time range of the determination target section is in a state of reduced consciousness. It is a specified value.
[0037]
The waveform continuation prescribing section width Td is a value for determining whether or not the section determined to be in a state of reduced consciousness continues. The moving cumulative total section set value nath is a value for setting a section in which the cumulative average of the moving average values is set.
[0038]
The determination condition setting means 15 is a means for setting determination conditions used by the traveling state determination means 14. The determination condition setting unit 15 sets a determination condition using information input via an input device (not shown) or a communication device (not shown) connected to the driving state determination device 1, the portable storage medium 16, and the like.
[0039]
The portable storage medium 16 is a readable / writable medium that stores various types of information stored in the in-vehicle storage device 13, and is realized by, for example, an IC card.
[0040]
Hereinafter, the processing operation of the driving state determination apparatus 1 will be described with reference to FIGS. Here, the driving state determination device 1 prescribes general roads and highways as road types, prescribes road type prescribing speed threshold values Vth for determining whether the road is a highway, and sets the vehicle speed value of the operating vehicle 2. When Vz exceeds the road type defining speed threshold Vth, it is determined that road type = highway.
[0041]
FIG. 2 shows an outline of the processing of the present invention.
[0042]
The moving average calculation unit 141 calculates various variables such as the moving average value Va and the difference value Vb from the vehicle speed value Vz in the vehicle speed data of the operating vehicle 2 measured by the vehicle speed sensor 11 and stored in the in-vehicle storage device 13 (step) S20).
[0043]
Then, the reduced consciousness state determination unit 142 refers to the determination condition, and determines whether or not the consciousness is lowered based on the calculated variables (moving average value Va, difference value Vb) (steps S21 to S24). ).
[0044]
That is, the reduced consciousness state determination unit 142 extracts points (poles) where the sign of the difference value Vb changes, and accumulates the points until there are three points (step S21). Thereafter, it is determined whether or not the three accumulated points satisfy the condition that defines the minimum interval of the consciousness lowered state (step S22).
[0045]
Further, it is determined whether or not a section (corresponding section) corresponding to the condition of three points is continuous within a predetermined time range with a corresponding section temporally preceding the section (step S23). Further, it is determined whether or not these consecutive condition-corresponding sections (continuous sections) correspond to a condition that defines a continuous state of reduced consciousness with high accuracy (step S24).
[0046]
If the three points do not meet the condition for defining the minimum section (step S22), it is determined that the section formed by the three points is normal travel (step S25).
[0047]
On the other hand, if the three points correspond to the above condition (step S22) but are not continuous with the corresponding section before that (step S23), the current corresponding section is defined as a single waveform section (first section). Determination is made, and the determination result is recorded in the in-vehicle storage device 13 (step S26).
[0048]
Furthermore, if the current relevant section is continuous with the previous relevant section (step S23), but the continuous relevant section does not correspond to the high-accuracy condition of the reduced consciousness state (step S24), the current relevant section A section formed by the section and the corresponding section before that is determined as a continuous waveform section (second section), and the determination result is recorded in the in-vehicle storage device 13 (step S27).
[0049]
On the other hand, when the current corresponding section is continuous with the previous corresponding section (step S23) and the continuous corresponding section group satisfies the high-accuracy condition of the reduced consciousness state (step S24), The section formed by the corresponding section group satisfying the condition is determined as a high-precision continuous waveform section (third section), and the determination result is recorded in the in-vehicle storage device 13 (step S28).
[0050]
FIG. 3 shows a more detailed process flow for the moving average calculation process in step S20.
[0051]
The road type determining means 12 acquires the vehicle speed value Vz (z = 1, 2, 3...) And time information of each vehicle speed value Vz sequentially measured from the vehicle speed sensor 11 (step S30), and a road type defining speed threshold value. The vehicle speed value Vz is determined based on Vth (step S31). When the vehicle speed value Vz exceeds the road type defining speed threshold value Vth, the processing after step S32 is performed.
[0052]
The moving average calculation unit 141 determines whether or not the data number na of the acquired vehicle speed value Vz has reached a predetermined moving cumulative total section set value nath at the present time (steps S32, S34, S38).
[0053]
First, if the data number na is smaller than the movement accumulation total section setting value nath −1 (step S32), the acquired vehicle speed value Vz is accumulated (step S33).
[0054]
If the data number na is equal to the movement accumulation total section setting value nath −1 (step S34), the acquired vehicle speed value Vz is accumulated (step S35), and thereafter the moving average of the vehicle speed values Vz accumulated up to the present time is obtained. The value Va is calculated (step S36), and then the difference value Vb (Vb = Va-Va-1) is calculated using the moving average value Va-1 calculated immediately before the vehicle speed value Vz (step S36). S37).
[0055]
If the number of data na is equal to the moving cumulative total section set value nath (step S38), the most recent data accumulated at the present time is discarded (step S39), and then from step S35 to step S37. Perform the process.
[0056]
Until the traveling of the operating vehicle 2 is completed (step S310), the processing from step S30 to step S39 is repeatedly executed every time the vehicle speed value Vz is input from the vehicle speed sensor 11. Also, in the consciousness lowered state determination process described below, the process is always performed using the latest moving average value Va and difference value Vb.
[0057]
4 and 5 show a more detailed processing flow for the consciousness lowered state determination processing in steps S21 to S28.
[0058]
The reduced consciousness state determination unit 142 acquires the moving average value Va and the difference value Vb from the moving average calculation unit 141 (step S40). At this time, it is checked whether the sign of the acquired difference value Vb and the immediately preceding difference value Vb-1 is inverted (step S41). The sign of the difference value Vb is the same as the sign of the immediately preceding difference value Vb-1. If they are equal, the subsequent determination process is not performed for the acquired moving average value Va and difference value Vb, and the process returns to step S40.
[0059]
On the other hand, if the sign of the difference value Vb is different from the sign of the immediately preceding difference value Vb-1, the acquired difference value Vb, its moving average value Va, and its acquisition time are used as the data for determining consciousness deterioration state. Further extraction is performed as (extraction point) (step S42).
[0060]
Subsequently, the number of extraction points currently extracted is determined (step S43). If the number of extraction points is not 3, the following determination processing is not performed, and the processing returns to step S40. On the other hand, if the number of extracted points is 3, the determination that defines a state of reduced consciousness with respect to the moving average value Va of the three extracted points (here, Va-2, Va-1, and Va are in time series order). Judgment is made using conditions.
[0061]
First, determination processing (step S44) of whether or not the three moving average values Va-2, Va-1, and Va satisfy the conditions "Va> Va-1 and Va-2> Va-1", and the difference Va It is determined whether -Va-2 satisfies a cruise state defining threshold value Vath that defines the cruise state in the moving average value change (step S45).
[0062]
If the conditions of steps S44 and S45 are satisfied, the difference (speed difference) Vl between the maximum moving average value (maximum speed) of the moving average values Va to Va-2 and the minimum moving average value (minimum speed) is calculated. (Step S46) Further, it is determined whether or not the difference Vl exceeds the abnormality regulation deceleration width Vthl that regulates the abnormal state of speed change (deceleration) (Step S47), and the minimum moving average value (minimum speed) is the vehicle. It is determined whether or not a driving condition defining threshold value Vlm that defines the normal running is exceeded (step S48).
[0063]
When the conditions of steps S47 and S48 are satisfied, the time range Tq of the three extraction points is calculated from the time information of the section (extraction section) formed by the three extraction points (step S49). It is determined whether or not Tt> Tq> Tb is satisfied with respect to the maximum time width Tt for lowering awareness and the minimum time width Tb for lowering awareness (step S410).
[0064]
In the determination processing from step S44 to step S410, if any of the conditions is not met, the data (Va-2) of the oldest extraction point in time series among the extraction is discarded (step S411), the subsequent processing is not performed, and the processing returns to step S40.
[0065]
On the other hand, in the determination processing from step S44 to step S410, when all the conditions are met, it is determined that the section of the three extraction points is a single waveform section (first section) that defines the state of consciousness reduction. Then, the determination result and the detection time are associated with each other, recorded in the in-vehicle storage device 13, and output from an output device such as sound and image (step S412). The determination result is output, for example, as audio data and notified to the driver of the operating vehicle 2, or is output as image data and displayed on a display device provided in the operating vehicle 2.
[0066]
After the completion of the processing in step S412, the oldest extraction point (Va-2) in time series among the extraction points is discarded (step S413), and the current extraction section determined to be in a state of reduced consciousness is the second in the same operation. It is determined whether or not it is a subsequent section (step S414). If the current extraction section is the first section, the process returns to step S40 without performing the subsequent processes.
[0067]
On the other hand, if the current extraction section is the second and subsequent sections, the extraction point (Va) that is the end of the extraction section (first section) that is in time series before the current extraction section and the beginning of the current extraction section. The time difference Ts with respect to a certain extraction point (Va-2) is calculated (step S415), and the time difference Ts is within the waveform continuation defining section width Td that defines the state in which a single waveform section is continuous (continuous waveform section). It is determined whether or not (step S416). If Ts <Td, the continuous waveform section (second second) in which the consciousness reduction operation state continues in the section from the beginning (extraction point) of the previous extraction section in the time series to the end (extraction point) of the current extraction section. (Interval), the determination result and the detection time are associated with each other, recorded in the in-vehicle storage device 13, and output from an output device such as voice or image (step S417). On the other hand, if Ts <Td, it is not determined as the second section, and the following process is not performed, and the process returns to step S40.
[0068]
Subsequently, with respect to the extracted section group determined to be the second section, the section time length Tqt of the (old) extracted section in the time series and the section time length Tqu of the (new) extracted section in the time series, and Then, it is determined whether or not the condition “Tqt <Tqu and Vlt <Vlu” is satisfied for the section speed difference Vlt of the extraction section before the time series and the section speed difference Vlu of the extraction section after the time series (step S418). .
[0069]
If the condition is met, the state of reduced consciousness is the period from the beginning (extraction point) of the last relevant section in the time series in the corresponding extraction section group to the end (extraction point) of the last relevant section. It is determined that the continuous state is a high-precision continuous waveform section (third section) that can be determined with high accuracy, the determination result and the detection time are associated with each other, recorded in the in-vehicle storage device 13, and output from an output device such as voice or image (Step S419). On the other hand, if the condition is not met, the third section is not determined, and the process returns to step S40.
[0070]
The above processing is repeatedly executed until the travel of the operating vehicle 2 is completed every time the moving average value Va and the difference value Vb are acquired (step S420).
[0071]
FIG. 6 shows an example of the traveling state of the operating vehicle 2 determined in the present invention. In FIG. 6, the vertical axis represents the moving average value Va (km / h), and the horizontal axis represents the time information (hh: mm) of the operation progress.
[0072]
With respect to the moving average value Va of the extraction points corresponding to the determination condition in step S41 of FIG. 4, the moving average of three consecutive extracted points that are always exceeding the running state defining threshold value Vlm and are the code change points of the moving average value Va. The value Va (Va, Va-1, Va-2) is the object of determination, and these moving average values Va (Va, Va-1, Va-2) are all equal to or greater than a certain abnormality regulating deceleration width Vthl, and The section (waveform) in which the time range Tq is within the range of the maximum time width Tt for lowering the state of consciousness and the minimum time width Tb for lowering the state of consciousness is classified and extracted from the single waveform section in the state of lowering consciousness. I understand that.
[0073]
The traveling state determination means 14 always uses the determined single waveform section (first section), continuous waveform section (second section), or high-precision continuous waveform section (third section) at all times. It is not necessary to record or output the section for each judgment. For example, it may be recorded or output only when an arbitrary section is detected, such as recording and outputting only the third section in which the state of reduced consciousness is determined with the highest accuracy.
[0074]
Furthermore, the output form of the three sections recorded as a result of the determination may be output after the fact using the recorded determination result information in addition to the real-time warning by voice or the like. In this case, the determination result stored in the in-vehicle storage device 13 is also recorded in the portable storage medium 16.
[0075]
FIG. 7 shows an output example when the determination result is output after the fact. In the output example of the determination result shown in FIG. 7, in the first section (P1), the time range from the start to the end of the single section corresponding to the waveform satisfying the determination condition of the consciousness reduction state is recorded. In the second section (P2) and the third section (P3), the time ranges from the start of the foremost first section to the end of the last first section when continuously generated are recorded. . As described above, the determination result of the driving state determination device 1, that is, the record of each section of the consciousness reduction state can be output in association with traveling vehicle speed information such as a digital operation recorder.
[0076]
As mentioned above, although this invention was demonstrated by the embodiment, this invention can be variously deformed in the range of the main point.
[0077]
For example, in the embodiment of the present invention, the driving state determination device 1 has been described as including the vehicle speed sensor 11. However, when the operating vehicle 2 equipped with the driving state determination device 1 includes a sensor capable of measuring the vehicle speed, the driving state determination device 1 does not need to include the vehicle speed sensor 11.
[0078]
Moreover, the driving | running state determination apparatus 1 may be connected to a communication processing means, and may transmit the information memorize | stored in the vehicle-mounted memory | storage device 13, or may receive the information used by the determination condition setting means 15. FIG.
[0079]
Further, the traveling state determination unit 14 of the driving state determination device 1 may be configured as an external processing mechanism of the traveling vehicle 2 and execute a determination process afterwards using information in the portable storage medium 16.
[0080]
The features of the embodiments and examples of the present invention are listed as follows.
[0081]
(Appendix 1) A processing method for determining a specific driving state of a vehicle,
Obtain vehicle speed information of the vehicle and acquisition time information of the vehicle speed,
Calculate a moving average value of an arbitrary section based on the vehicle speed information, calculate a moving average difference value between the moving average value and the moving average value immediately before the moving average value,
The moving average value and the moving average difference value are stored as determination data in association with the acquisition time information,
Among the determination data, the moving average difference value having a sign different from the moving average difference value immediately before the moving average difference value is accumulated as an extraction point.
The extraction is performed for a section in which the sign of the moving average difference value of the extraction points includes an extraction point converted from positive to negative, an extraction point converted from negative to positive, and an extraction point converted from positive to negative. The time length of the section is in an arbitrary time range, and the average speed difference in the section, which is the difference between the maximum moving average value and the minimum moving average value in the section, is in an arbitrary deceleration range, and the moving average value When the difference between the moving average values of the two extraction points whose sign of the difference value has changed from positive to negative is within an arbitrary speed range, the section is determined as a single waveform section indicating a state of reduced consciousness,
Output the determination result or information used for the determination
The driving | running state determination processing method characterized by the above-mentioned.
[0082]
(Additional remark 2) It is the driving | running state determination processing method of the said additional remark 1, Comprising:
When the single waveform section occurs within an arbitrary time range with a single waveform section that is in time series before the single waveform section, the start of the single waveform section that is in front of the time series To the end of the single waveform section is determined as a continuous waveform section indicating a state where the state of consciousness continues,
Output the determination result or information used for the determination
The driving | running state determination processing method characterized by the above-mentioned.
[0083]
(Additional remark 3) It is the driving | running state determination processing method of the said additional remark 2, Comprising:
In the single waveform section group constituting the continuous waveform section, the values of both the time length of the single waveform section and the average speed difference within the section in the time series are in front of the single waveform section. When the time length of the single waveform section is larger than the time difference and the speed difference in the section, from the beginning of the single waveform section in the forefront of the time series to the end of the single waveform section in the end of the time series in the single waveform section group Is determined as a high-precision continuous waveform section indicating a high-precision state in which the state of reduced consciousness continues,
Output the determination result or information used for the determination
The driving | running state determination processing method characterized by the above-mentioned.
[0084]
(Appendix 4) A processing device for determining a specific driving state of a vehicle,
Vehicle speed information acquisition means for acquiring vehicle speed information of the vehicle and acquisition time information of the vehicle speed;
A moving average calculating means for calculating a moving average value of an arbitrary section based on the vehicle speed information, and calculating a moving average difference value between the moving average value and a moving average value immediately before the moving average value;
Storage means for holding the moving average value and the moving average difference value as determination data in association with the acquisition time information;
Among the data for determination, the sign of the moving average difference value that is different from the sign of the moving average difference value immediately before the moving average difference value is accumulated as an extraction point, and the sign of the moving average difference value of the extraction point is , For an interval composed of an extraction point converted from positive to negative, an extraction point converted from negative to positive, and an extraction point converted from positive to negative, the time length of the extraction interval is within an arbitrary time range. Yes, and the average speed difference in the section, which is the difference between the maximum moving average value and the minimum moving average value in the section, is within an arbitrary deceleration range, and the sign of the moving average value difference value has changed from positive to negative When the difference between the moving average values of the two extraction points is within an arbitrary speed range, the section is determined as a single waveform section indicating a state of reduced consciousness, and the determination result or information used for the determination is output. To determine the state of consciousness Prepare for
The driving | running state determination apparatus characterized by the above-mentioned.
[0085]
(Additional remark 5) It is a program for making a computer perform the process which determines the specific driving state of a vehicle,
Processing for acquiring vehicle speed information of the vehicle and acquisition time information of the vehicle speed;
A process of calculating a moving average value of an arbitrary section based on the vehicle speed information, and calculating a moving average difference value between the moving average value and a moving average value immediately before the moving average value;
A process of associating the moving average value and the moving average difference value with the acquisition time information and holding as determination data;
A process of accumulating, as an extraction point, the data for determination in which the sign of the moving average difference value is different from the sign of the moving average difference value immediately before the moving average difference value,
The extraction is performed for a section in which the sign of the moving average difference value of the extraction points includes an extraction point converted from positive to negative, an extraction point converted from negative to positive, and an extraction point converted from positive to negative. The time length of the section is in an arbitrary time range, and the average speed difference in the section, which is the difference between the maximum moving average value and the minimum moving average value in the section, is in an arbitrary deceleration range, and the moving average value A process of determining the section as a single waveform section indicating a state of reduced consciousness when the difference between the moving average values of the two extraction points whose sign of the difference value has changed from positive to negative is within an arbitrary speed range; ,
A process of outputting the determination result or the information used for the determination,
Let computer run
An operating state determination processing program characterized by the above.
[0086]
(Additional remark 6) The recording medium which recorded the program for making a computer perform the process which determines the specific driving | running state of a vehicle,
Processing for acquiring vehicle speed information of the vehicle and acquisition time information of the vehicle speed;
A process of calculating a moving average value of an arbitrary section based on the vehicle speed information, and calculating a moving average difference value between the moving average value and a moving average value immediately before the moving average value;
A process of associating the moving average value and the moving average difference value with the acquisition time information and holding as determination data;
A process of accumulating, as an extraction point, a data in which the sign of the moving average difference value is different from the sign of the moving average difference value immediately before the moving average difference value among the determination data;
The extraction is performed for a section in which the sign of the moving average difference value of the extraction points includes an extraction point converted from positive to negative, an extraction point converted from negative to positive, and an extraction point converted from positive to negative. The time length of the section is in an arbitrary time range, and the average speed difference in the section, which is the difference between the maximum moving average value and the minimum moving average value in the section, is in an arbitrary deceleration range, and the moving average value A process of determining the section as a single waveform section indicating a state of reduced consciousness when the difference between the moving average values of the two extraction points whose sign of the difference value has changed from positive to negative is within an arbitrary speed range; ,
A process of outputting the determination result or the information used for the determination,
An operating state determination processing program recording medium for recording a processing program to be executed by a computer.
[0087]
【The invention's effect】
As described above, according to the present invention, it becomes possible to detect a reduced consciousness state even in an environment with a low load or a low storage capacity, and the consciousness can be detected with a general-purpose in-vehicle information processing mechanism or a portable information terminal. A lowered state determination process can be realized.
[0088]
Further, according to the present invention, it is possible to realize a method for determining a state of consciousness reduction immediately on a vehicle at a lower cost, and the economics of operation management is improved.
[Brief description of the drawings]
FIG. 1 is a diagram illustrating a configuration example of a processing apparatus that implements the present invention.
FIG. 2 is a diagram showing an outline of processing according to the present invention.
FIG. 3 is a diagram illustrating a detailed processing flow of a moving average calculation process.
FIG. 4 is a diagram showing a detailed processing flow of a consciousness lowered state determination process.
FIG. 5 is a diagram illustrating a detailed processing flow of a consciousness lowered state determination process;
FIG. 6 is a diagram showing an example of a traveling state of a running vehicle determined in the present invention.
FIG. 7 is a diagram illustrating an output example when a determination result is output after the fact.
[Explanation of symbols]
1 Driving state determination device (on-vehicle device)
11 Vehicle speed sensor
12 Road type determination means
13 On-vehicle storage device
14 Traveling state determination means
141 Moving average calculator
142 Degradation state determination unit
15 Judgment condition setting means
16 Portable storage media
17 Judgment results (sound, image, etc.)
2 Operating vehicles

Claims

A processing method for determining a driver 's consciousness reduction state while driving a vehicle,
Obtain vehicle speed information of the vehicle and acquisition time information of the vehicle speed information ,
Calculate a moving average value of a predetermined section based on the vehicle speed information, calculate a moving average difference value between the moving average value and the moving average value immediately before the moving average value,
The moving average value and the moving average difference value are stored as determination data in association with the acquisition time information,
Among the determination data, the moving average difference value having a sign different from the moving average difference value immediately before the moving average difference value is accumulated as an extraction point.
Sign of the moving average differential value of the extraction point, extraction points converted from positive to negative, and extracted extracted points converted from negative to positive, and a section consisting of a sequence of conversion was extracted point from positive to negative, A section in which the time length of the extracted section is within a minimum time range indicating a minimum section in a predetermined state of reduced consciousness and is a difference between the maximum moving average value and the minimum moving average value in the extracted section When the average average speed difference is equal to or greater than a threshold value indicating a predetermined abnormality regulation deceleration width and the minimum speed in the extracted segment is equal to or greater than a predetermined running condition regulation threshold value , the extracted interval is Judged as a single waveform section indicating a state of consciousness decline,
When the interval between the single waveform section and the single waveform section that is in time series before the single waveform section is within a predetermined interval determination time range, the single waveform section in front of the time series The section from the beginning of one waveform section to the end of the single waveform section is determined as a continuous waveform section indicating a state in which the state of reduced consciousness continues,
The determination result or the information used for the determination is output. A driving state determination processing method.

It is the driving | running state determination processing method of Claim 1, Comprising:
In the single waveform section group constituting the continuous waveform section, the values of both the time length of the single waveform section and the average speed difference within the section in the time series are in front of the single waveform section. When the time length of the single waveform section is larger than the time difference and the speed difference in the section, from the beginning of the single waveform section in the forefront of the time series to the end of the single waveform section in the end of the time series in the single waveform section group the interval, determines that the high-accuracy continuous waveform section showing a state in which drowsiness is continuous with high precision,
The determination result or the information used for the determination is output. A driving state determination processing method.

A processing device for determining a driver 's consciousness reduction state while driving a vehicle,
A vehicle speed information obtaining means for obtaining the acquisition time information of the vehicle speed information and the vehicle speed information of the vehicle,
A moving average calculating means for calculating a moving average value of a predetermined section based on the vehicle speed information, and calculating a moving average difference value between the moving average value and a moving average value immediately before the moving average value;
Storage means for holding the moving average value and the moving average difference value as determination data in association with the acquisition time information;
Among the data for determination, the sign of the moving average difference value that is different from the sign of the moving average difference value immediately before the moving average difference value is accumulated as an extraction point, and the sign of the moving average difference value of the extraction point is , extracted points converted from positive to negative, extracted points converted from negative to positive, and a sequence of positive from converted to the negative extraction point to extract composed interval, the time length of the extracted segment is predetermined An abnormal regulation in which an average speed difference in a section that is within a minimum time range indicating a minimum section that is in a state of reduced consciousness and that is a difference between the maximum moving average value and the minimum moving average value in the extracted section is predetermined. When the minimum speed in the extracted section is equal to or greater than a predetermined threshold for driving condition definition , the extracted section is determined as a single waveform section indicating a state of reduced consciousness. and, further When the interval between the single waveform section and the single waveform section that is in time series before the single waveform section is within a predetermined interval determination time range, the single waveform section in front of the time series Decrease consciousness state determination in which a section from the beginning of one waveform section to the end of the single waveform section is determined as a continuous waveform section indicating a state in which the state of consciousness continues, and the determination result or information used for the determination is output An operation state determination device comprising: means.

The operation state determination device according to claim 3,
In the single waveform section group that constitutes the continuous waveform section, both the time length of the single waveform section and the value of the average speed difference in the section that are after the time series are in the time series. When the time length of the previous single waveform section and the speed difference within the section are larger, the single waveform section group is located at the end of the time series from the beginning of the single waveform section at the forefront of the time series. The section up to the end of the single waveform section is determined as a high-precision continuous waveform section that indicates the state in which the state of reduced consciousness continues with high accuracy
The driving | running state determination apparatus characterized by the above-mentioned.

In order to determine the driver 's consciousness reduction state while driving the vehicle ,
Processing for acquiring vehicle speed information of the vehicle and acquisition time information of the vehicle speed information ;
A process of calculating a moving average value of a predetermined section based on the vehicle speed information, and calculating a moving average difference value between the moving average value and a moving average value immediately before the moving average value;
A process of associating the moving average value and the moving average difference value with the acquisition time information and holding as determination data;
A process of accumulating, as an extraction point, the data for determination in which the sign of the moving average difference value is different from the sign of the moving average difference value immediately before the moving average difference value,
Sign of the moving average differential value of the extraction point, extraction points converted from positive to negative, and extracted extracted points converted from negative to positive, and a section consisting of a sequence of conversion was extracted point from positive to negative, A section in which the time length of the extracted section is within a minimum time range indicating a minimum section that is in a predetermined state of reduced consciousness , and is a difference between the maximum moving average value and the minimum moving average value in the extracted section When the average average speed difference is equal to or greater than a threshold value indicating a predetermined abnormality regulation deceleration width and the minimum speed in the extracted segment is equal to or greater than a predetermined running condition regulation threshold value , the extracted interval is A process of determining a single waveform section indicating a state of reduced consciousness;
When the interval between the single waveform section and the single waveform section that is in time series before the single waveform section is within a predetermined interval determination time range, the single waveform section in front of the time series A process for determining a continuous waveform section indicating a state in which the state of reduced consciousness is continuous from the beginning of one waveform section to the end of the single waveform section;
A process of outputting the determination result or the information used for the determination,
Operating state determination processing program for execution.