JP4563619B2

JP4563619B2 - Vehicle driving support device

Info

Publication number: JP4563619B2
Application number: JP2001196836A
Authority: JP
Inventors: 弘幸関口
Original assignee: Fuji Jukogyo KK
Current assignee: Subaru Corp
Priority date: 2001-06-28
Filing date: 2001-06-28
Publication date: 2010-10-13
Anticipated expiration: 2021-06-28
Also published as: JP2003016593A

Description

【０００１】
【発明の属する技術分野】
本発明は、自車両が車線を逸脱して走行する可能性を適切に判断し、車線逸脱の可能性が有る際には警報を発してドライバの運転を支援する車両用運転支援装置に関する。
【０００２】
【従来の技術】
近年、車両の安全性の向上を図るため、積極的にドライバの運転操作を支援する総合的な運転支援システム（ＡＤＡ；Active Drive Assist system）が開発されている。このＡＤＡシステムは、車両の走行環境情報や自車両の走行状態から先行車両との衝突、物体との接触、車線逸脱等の様々な可能性を推定して、安全を維持できないと予測される場合に、ドライバに対して報知、その他制御等を行なうものである。
【０００３】
上記ＡＤＡシステムの機能の一つである車線逸脱防止は、車線に極度に接近したり逸脱したりすると、その事を警報等でドライバに知らせ、ドライバのぼんやりや居眠りやよそ見等の不注意で生じる交通事故の防止を図るもので、この車線逸脱防止機能については既に様々な提案がなされている。例えば、特開平９−１４２３２７号公報では、自車両前方の映像から自車両が現在走行している走行レーンを判定し、また、近い将来の自車両の走行軌跡を推定して、走行レーンと走行軌跡とを重ね合わせることにより、自車両が現在の走行レーンを逸脱する可能性の有無を判定する技術が開示されている。
【０００４】
【発明が解決しようとする課題】
ところで、上述のような車線逸脱防止では、自車両の走行軌跡の推定が非常に重要であるが、操舵角やヨーレートを基に走行軌跡を推定する場合、カーブ進入時では、操舵が十分に行われておらず、またヨーレートの発生も少ないため、推定する走行軌跡の走行曲率が十分に精度良く得られず、自車両が現在の走行レーンを逸脱すると判定される場合が多いという問題があった。また、Ｓ字カーブ等の切り返しでは、実際の自車両の動きに対して演算される走行曲率が遅れることから、やはり自車両が現在の走行レーンを逸脱すると判定される場合が多かった。
【０００５】
本発明は上記事情に鑑みてなされたもので、車線逸脱の可能性を精度良く判断することができ、自車両が現在の走行レーンを逸脱すると判定される場合が不必要に増加することを防止し、自然で使い良い車両用運転支援装置を提供することを目的としている。
【０００６】
【課題を解決するための手段】
上記目的を達成するため請求項１記載の本発明による車両用運転支援装置は、自車両前方の走行路を認識する走行路認識手段と、上記自車両の走行軌跡を推定する走行軌跡推定手段と、上記走行路と上記走行軌跡とを比較して上記自車両の上記走行路からの逸脱を判定する逸脱判定手段とを備えた車両用運転支援装置において、上記逸脱判定手段は、上記自車両の走行状態がカーブ進入の状態にあると判定した際は、上記走行路と上記走行軌跡とを比較して上記自車両の上記走行路からの逸脱を判定して第１の判定結果とすると共に、上記走行軌跡推定手段で推定した走行軌跡を補正し、この補正した走行軌跡と上記走行路とを比較して上記自車両の上記走行路からの逸脱を判定して第２の判定結果とし、上記第１の判定結果と上記第２の判定結果とにより上記自車両の上記走行路からの逸脱を判定することを特徴としている。
【０００７】
すなわち、上記請求項１記載の車両用運転支援装置は、走行路認識手段で自車両前方の走行路を認識し、走行軌跡推定手段で自車両の走行軌跡を推定し、逸脱判定手段で走行路と上記走行軌跡とを比較して上記自車両の上記走行路からの逸脱を判定する。この逸脱判定で、逸脱判定手段は、自車両の走行状態がカーブ進入の状態にあると判定した際は、走行路と走行軌跡とを比較して自車両の走行路からの逸脱を判定して第１の判定結果とすると共に、走行軌跡推定手段で推定した走行軌跡を補正し、この補正した走行軌跡と走行路とを比較して自車両の走行路からの逸脱を判定して第２の判定結果とし、第１の判定結果と第２の判定結果とにより自車両の走行路からの逸脱を判定する。このように自車両の走行状態がカーブ進入の状態にあると判定した際は、第１の判定結果と第２の判定結果とにより自車両の走行路からの逸脱を判定するようにしたので、車線逸脱の可能性を精度良く判断することができ、自車両が現在の走行レーンを逸脱すると判定される場合が不必要に増加することが防止され、自然で使い良い装置となる。
【０００８】
また、請求項２記載の本発明による車両用運転支援装置は、請求項１記載の車両用運転支援装置において、上記逸脱判定手段は、上記第１の判定結果と上記第２の判定結果の両方が、上記自車両が上記走行路から逸脱するとなった場合に上記自車両が上記走行路から逸脱すると判定することを特徴とし、すなわち、どちらかのみが自車両が走行路から逸脱するとの結果となっても、この結果はキャンセルされて自車両が現在の走行レーンを逸脱すると判定される場合が不必要に増加することが防止される。
【０００９】
更に、請求項３記載の本発明による車両用運転支援装置は、請求項１又は請求項２に記載の車両用運転支援装置において、上記逸脱判定手段は、転舵の速さと自車速に応じて上記自車両の走行状態がカーブ進入の状態にあるか否かを判定することを特徴とし、例えば直線路や高速道路で観られる車両挙動の場合では自車両の走行状態がカーブ進入の状態にあると判定しないようにする。
【００１０】
また、請求項４記載の本発明による車両用運転支援装置は、請求項１乃至請求項３の何れか一つに記載の車両用運転支援装置において、上記逸脱判定手段は、上記第２の判定結果を得るために用いる上記補正した走行軌跡は、上記走行軌跡推定手段で推定する走行曲率の大きさを自車速と操舵角に応じて補正した値で求めることを特徴とし、例えばカーブへの進入速度が高く操舵角が小さいほど先のカーブが緩いと判断できるので、自車速に反比例、操舵角に比例した修正を行う。
【００１１】
更に、請求項５記載の本発明による車両用運転支援装置は、請求項１乃至請求項４の何れか一つに記載の車両用運転支援装置において、上記逸脱判定手段は、上記第２の判定結果を得るために用いる上記補正した走行軌跡は、上記走行軌跡推定手段で推定する走行曲率の方向を転舵方向に応じて補正した値で求めることを特徴とし、例えば転舵の方向と走行曲率の方向が同じになるように走行曲率の方向を補正し、Ｓ字カーブでの切り返しなどでの誤判定を低減する。
【００１２】
また、請求項６記載の本発明による車両用運転支援装置は、請求項１乃至請求項５の何れか一つに記載の車両用運転支援装置において、上記逸脱判定手段は、上記第１の判定結果を得るために用いる上記走行軌跡は所定の平均走行曲率で求めると共に、上記第２の判定結果を得るために用いる上記補正した走行軌跡は現在推定した走行曲率を補正して求めることを特徴とし、第２の判定結果には、現在推定した走行曲率を補正して求めることで、実際の自車両の動きに対する走行曲率を求めることの応答性を向上し、Ｓ字カーブ等での迅速で正確な判定に寄与する。
【００１３】
【発明の実施の形態】
以下、図面に基づいて本発明の実施の形態を説明する。
図１〜図３は本発明の実施の形態を示し、図１は車両用運転支援装置の概略構成図、図２は車線逸脱判断の説明図、図３は車線逸脱防止制御のフローチャートである。
【００１４】
図１において、符号１は自動車等の車両（自車両）であり、この自車両１に、車線に極度に接近したり逸脱したりすると、その事を警報等でドライバに知らせ、ドライバの居眠りやよそ見等の不注意で生じる交通事故の防止を図る、いわゆる車線逸脱防止機能を一つの機能として有する車両用運転支援装置（ＡＤＡ装置）２が搭載されている。以下、本発明の実施の形態では、ＡＤＡ装置２の車線逸脱防止機能の部分についてのみ説明し、他の機能の部分については説明を省略する。
【００１５】
ＡＤＡ装置２は、ステレオ光学系として例えば電荷結合素子（ＣＣＤ）等の固体撮像素子を用いた１組の（左右の）ＣＣＤカメラ３を有し、これら左右のＣＣＤカメラ３は、それぞれ車室内の天井前方に一定の間隔をもって取り付けられ、車外の対象を異なる視点からステレオ撮像するようになっている。
【００１６】
また、車両１には、車速Ｖを検出する車速センサ４、ヨーレートγを検出するヨーレートセンサ５、及び操舵角δを検出する操舵角センサ６が設けられており、これら検出された自車速Ｖ、ヨーレートγ、操舵角δ及び１組のＣＣＤカメラ３で撮像した自車両１の走行方向の画像は、制御装置７に入力される。
【００１７】
制御装置７は、自車両１の走行方向の画像と自車速Ｖ、ヨーレートγ、操舵角δが入力され、自車両１の走行路前方の車線逸脱の可能性を判断して、自車両１が車線逸脱の可能性が有ると判断した際に、コンビネーションメータ８の警報ランプ９と警報ブザー１０とを作動させる。
【００１８】
すなわち、制御装置７は、イメージプロセッサを有するマルチマイクロプロセッサのシステムで形成され、１組のＣＣＤカメラ３で撮像した１組のステレオ画像対に対し、対応する位置のずれ量から三角測量の原理によって画像全体に渡る距離情報を求める処理を行なって、三次元の距離分布を表す距離画像を生成する。そして、この距離画像を処理して必要な情報を抽出して、この情報を基に自車両１の走行路前方の車線逸脱の可能性を判断する。
【００１９】
具体的には、距離画像から車線、車両、車両以外の物体等を分離して検出し、道路、物体の画像データを得る。この画像データにより、車線については、例えば、図２に示すように、自車両１の位置と方位を基準とするＸ−Ｙ座標（車両の幅の略中央で略バンパ先端位置を原点Ｏとする座標）において、車両前方の左右の車線の位置や形状の座標等に関するデータを得る。
【００２０】
その一方で、制御装置７は、現在の自車速Ｖとヨーレートγ、或いは、自車速Ｖと操舵角δにより、自車両１の走行軌跡、すなわち旋回曲率と自車両１の幅寸法（例えば１．７ｍ）で得られる軌跡を推定する。そして、この推定した旋回曲率で走行した際の自車両１前方の第１の距離（例えば、前方約１０．２４ｍ）に自車両１の幅方向と平行に第１の判断ラインＬK を設定し、自車両１前方の第２の距離（例えば、前方約１８．４３２ｍ）に自車両の幅方向と平行な第２の判断ラインＬE を設定する。尚、前方約１０．２４ｍと前方約１８．４３２ｍにおける自車両１の位置は、予め旋回曲率を変数とする関数で与えられる。こうして、制御装置７は、第１の判断ラインＬK での自車両位置と車線位置とに応じて車線逸脱の可能性を判断し、第２の判断ラインＬE での自車両位置と車線位置とに応じて車線逸脱の可能性を判断して、これら２つの判断結果が共に車線逸脱の可能性有り（走行車線が自車両１の推定された走行軌跡内に重なる場合）と判断した場合に自車両１は車線逸脱の可能性有りと最終的に判断する。ここで、第１の判断ラインＬK 及び第２の判断ラインＬE における自車両位置は、走行車線の左側車線が自車両１の推定された走行軌跡内に重なる場合は左側に車線逸脱、或いは、走行車線の右側車線が自車両１の推定された走行軌跡内に重なる場合は右側に車線逸脱と判断する。
【００２１】
また、制御装置７は、上述の車線逸脱判定を、後述の車線逸脱防止制御のフローチャートで詳述する如く、自車両１の走行状態がカーブ進入の状態にあるか否かで分けて行うようになっている。具体的には、自車速Ｖと転舵速度（操舵角δの微分値）とから自車両１の走行状態がカーブ進入の状態にあるか否か判定し、例えば自車速Ｖが３０km/h〜６０km/hで、且つ、転舵速度が４０deg/s以上の場合はカーブ進入時と判定する。
【００２２】
そして、自車両１の走行状態が上述の自車速Ｖと転舵速度の条件を満足せず、直進路或いは高速道路を走行中と判断できる場合は、過去０．３秒間の平均旋回曲率Ｃu3を基に上述の車線逸脱判定を行う（第１の判定結果）。
【００２３】
一方、自車両１の走行状態が上述の自車速Ｖと転舵速度の条件を満足し、カーブ進入の状態にあると判定した場合は、現在演算されている旋回曲率Ｃuaの大きさ、方向を修正（補正）し、この修正した旋回曲率Ｃuaを基に上述の車線逸脱判定を行う（第２の判定結果）。
【００２４】
こうして、自車両１の走行状態がカーブ進入の状態に無いと判定した場合は第１の判定結果を最終的な判定結果とする一方、自車両１の走行状態がカーブ進入の状態にあると判定した場合は第１の判定結果と第２の判定結果とを比較して、これらが共に車線逸脱と判定される場合にのみ最終的な判定結果は車線逸脱とされる。
【００２５】
以上の如く制御装置７は、本実施の形態においては、走行路認識手段、走行軌跡推定手段、逸脱判定手段としての機能を備えている。
【００２６】
また、制御装置７は、上述の車線逸脱の可能性の判断結果に基づき、自車両１が車線逸脱の可能性が有る場合には警報を発する。この警報は、例えば、車速に応じて行われ、車速Ｖが４０km/h〜５０km/hまでは警報ランプ９のみの警報が行われ、車速Ｖが５０km/h以上では警報ランプ９と警報ブザー１０の両方による警報を行うように制御される。
【００２７】
次に、上記構成によって達成される車線逸脱防止の機能を、図３に示す車線逸脱防止制御のフローチャートで説明する。この車線逸脱防止制御のプログラムは、所定時間、例えば１００ms毎に繰り返し実行され、まず、ステップ（以下、「Ｓ」と略称）１０１で、画像認識処理が実行され、前述の如く、１組のＣＣＤカメラ３で撮像した自車両１の走行方向の画像を基に、前方の左側車線、右側車線を認識する。
【００２８】
その後、Ｓ１０２に進み、ヨーレートγ、自車速Ｖ、操舵角δ等の必要なパラメータを読み込み、Ｓ１０３に進む。
【００２９】
Ｓ１０３では、ヨーレートセンサ５が有効か否か判定し、ヨーレートセンサ５が有効であれば、Ｓ１０４に進みヨーレートγと自車速Ｖを用いて例えば以下（１）式により現在の旋回曲率Ｃuaが算出される。
Ｃua＝γ／Ｖ …（１）
【００３０】
一方、Ｓ１０３でヨーレートセンサ５が有効ではないと判定された場合はＳ１０５に進み、操舵角δが所定値（例えば０．５７度）以上で転舵が行われているか否か判定される。
【００３１】
Ｓ１０５での判定の結果、操舵角δが０．５７度以上で操舵が行われていると判定できる場合はＳ１０６に進み、操舵角δと自車速Ｖを用いて例えば以下（２）、（３）式により現在の旋回曲率Ｃuaが算出される。
Ｒｅ＝（１＋Ａ・Ｖ^２）・（Ｌ／δ） …（２）
Ｃua＝１／Ｒｅ …（３）
ここで、Ｒｅは旋回半径、Ａは車両のスタビリティファクタ、Ｌはホイールベースである。
【００３２】
また、Ｓ１０５での判定の結果、操舵角δが０．５７度未満で操舵が行われていると判定できる場合はＳ１０７に進み、現在の旋回曲率Ｃuaは０に設定される。
【００３３】
Ｓ１０４、Ｓ１０６或いはＳ１０７で現在の旋回曲率Ｃuaを設定した後は、Ｓ１０８に進み、この現在の旋回曲率Ｃuaを加えた過去約０．３秒間の旋回曲率から平均旋回曲率Ｃu3を算出する。
【００３４】
その後、Ｓ１０９に進み、平均旋回曲率Ｃu3を基に第１の判断ライン（１０．２４ｍ）、第２の判断ライン（１８．４３２ｍ）の自車位置を推定し、白線位置との比較にて逸脱判定を行い、この判定結果を第１の判定結果とする。すなわち、第１の判断ライン、第２の判断ラインの何れにおいても自車両１の走行軌跡内に車線が重なる場合にのみ車線逸脱と判定する。
【００３５】
次いで、Ｓ１１０に進み、自車両１の走行状態がカーブ進入の状態にあるか否か、すなわち、自車速Ｖが３０km/h〜６０km/hで、且つ、転舵速度が４０deg/s以上の場合か否かを判定し、自車両１の走行状態が上述の自車速Ｖと転舵速度の条件を満足せず、直進路或いは高速道路を走行中と判断できる場合はＳ１１１に進み、第２の判定結果に第１の判定結果を代入して、これを第２の判定結果とする。
【００３６】
一方、Ｓ１１０で、自車両１の走行状態が上述の自車速Ｖと転舵速度の条件を満足し、カーブ進入の状態にあると判定した場合は、Ｓ１１２に進み、現在の旋回曲率Ｃuaの大きさを修正する。具体的には、現在の旋回曲率Ｃuaに乗算する修正係数ｋを、以下の（４）式に示すように設定する。
【００３７】
ｋ＝｜α・（δ／Ｖ）｜ …（４）
ここで、αはゲインである。すなわち、例えばカーブへの進入速度が高く操舵角が小さいほど先のカーブが緩いと判断できるので、自車速Ｖに反比例、操舵角δに比例した修正を行うのである。また、ゲインαは、第１の判断ライン（１０．２４ｍ）用と、第２の判断ライン（１８．４３２ｍ）用で用意され、第１の判断ライン（１０．２４ｍ）では小さい値に、第２の判断ライン（１８．４３２ｍ）では大きい値に設定され、曲がり込んでいるカーブに合わせた走行曲率の修正が行われるようになっている。そして、このように現在の旋回曲率Ｃuaを補正することにより、自車両１のカーブ進入時に、たとえ操舵角δ、ヨーレートγが小さい値であっても、推定される現在の旋回曲率Ｃuaが正確な値に補正されるのである。
【００３８】
その後、Ｓ１１３に進み、現在の旋回曲率Ｃuaの方向を修正する。具体的には、現在の旋回曲率Ｃuaの方向と操舵方向とを比較して、一致しない場合、現在の旋回曲率Ｃuaに乗算する修正係数ｋを、ｋ＝−ｋとしてその方向を操舵方向と一致するように補正する。すなわち、Ｓ字カーブでの切り返しなどで、ヨーレートγが操舵方向に十分に現れない場合を考慮し、これを補正するのである。
【００３９】
次いで、Ｓ１１４に進み、上述のＳ１１２及びＳ１１３で修正（補正）した修正係数ｋを現在の旋回曲率Ｃuaに乗算し、この補正した現在の旋回曲率Ｃuaを基に第１の判断ライン（１０．２４ｍ）、第２の判断ライン（１８．４３２ｍ）の自車位置を推定し、白線位置との比較にて逸脱判定を行い、この判定結果を第２の判定結果とする。すなわち、第１の判断ライン、第２の判断ラインの何れにおいても自車両１の走行軌跡内に車線が重なる場合にのみ車線逸脱と判定する。このように第２の判定結果は、補正した現在の旋回曲率Ｃuaを基に得られるので、左右のカーブの連続した走行路であってもレスポンス良く正確な判定結果となる。
【００４０】
Ｓ１１１或いはＳ１１４で第２の判定結果を得た後は、Ｓ１１５に進み、第１の判定結果と第２の判定結果の比較が行われる。そして、この比較の結果、２つの逸脱判定結果が共に逸脱の場合はＳ１１６に進み、逸脱警報を発生させ、それ以外の場合はＳ１１７に進み、逸脱警報を行わず、ルーチンを抜ける。
【００４１】
そして、Ｓ１１６で行われる逸脱警報は、上述の如く、例えば、車速に応じて行われ、車速Ｖが４０km/h〜５０km/hまでは警報ランプ９のみの警報が行われ、左側への車線逸脱か、或いは、右側への車線逸脱か、も警報ランプ９の点灯で報知される。また、車速Ｖが５０km/h以上では警報ランプ９に加え、警報ブザー１０によっても警報が行われる。
【００４２】
このように、本実施の形態によれば、自車両１の走行状態がカーブ進入の状態にあると判定した際は、車線と走行軌跡とを比較して自車両１の車線からの逸脱を判定して第１の判定結果とすると共に、推定した現在の走行軌跡を補正し、この補正した現在の走行軌跡と車線とを比較して自車両１の車線からの逸脱を判定して第２の判定結果とし、第１の判定結果と第２の判定結果とにより自車両の車線からの逸脱を判定する。すなわち、図２に示すように、自車両１の走行状態がカーブ進入の状態にあると判定した際は、従来では図中のＳk1で与えられていた走行軌跡が、現在の走行軌跡を補正してＳk2で与えられる。このため、従来の走行軌跡Ｓk1では、第１の判断ラインＬKにおいて走行軌跡上の点ＰKL1より内側に車線上の点ＰKL0が位置し、第２の判断ラインＬＥにおいて走行軌跡上の点ＰEL1より内側に車線上の点ＰEL0が位置して車線逸脱と誤判定される。これに対し本実施の形態では、第１の判断ラインＬKにおいて走行軌跡上の点ＰKL2より外側に車線上の点ＰKL0が位置し、第２の判断ラインＬＥにおいて走行軌跡上の点ＰEL2より外側に車線上の点ＰEL0が位置して車線逸脱せずと第２の判定結果で正確に判定される。これにより本実施の形態では最終的に車線逸脱せずと判定され、たとえカーブ進入時等でヨーレートγや操舵角δが小さい場合でも正確な車線逸脱判定が行え、自車両１が現在の走行レーンを逸脱すると判定される場合が不必要に増加することが防止され、自然で使い良い装置となっている。
【００４３】
【発明の効果】
以上説明したように本発明によれば、車線逸脱の可能性を精度良く判断することができ、自車両が現在の走行レーンを逸脱すると判定される場合が不必要に増加することを防止し、自然で使い良いという効果を得ることができる。
【図面の簡単な説明】
【図１】車両用運転支援装置の概略構成図
【図２】車線逸脱判断の説明図
【図３】車線逸脱防止制御のフローチャート
【符号の説明】
１自車両
２車両用運転支援装置
３ＣＣＤカメラ
７制御装置（走行路認識手段、走行軌跡推定手段、逸脱判定手段）[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a vehicle driving support device that appropriately determines the possibility that a host vehicle deviates from a lane, and issues a warning to assist the driving of a driver when there is a possibility of lane departure.
[0002]
[Prior art]
In recent years, in order to improve the safety of vehicles, a comprehensive driving assistance system (ADA) that actively supports driving operation of a driver has been developed. This ADA system estimates various possibilities such as a collision with a preceding vehicle, contact with an object, lane departure, etc. from the traveling environment information of the vehicle and the traveling state of the host vehicle, and it is predicted that safety cannot be maintained. In addition, notification and other control are performed for the driver.
[0003]
Lane departure prevention, which is one of the functions of the ADA system, occurs when the driver approaches or departs from the lane extremely, alerting the driver with an alarm, etc., and inadvertently causing the driver to fall asleep or snooze or look away. In order to prevent traffic accidents, various proposals have already been made for this lane departure prevention function. For example, in Japanese Patent Application Laid-Open No. 9-142327, a traveling lane in which the host vehicle is currently traveling is determined from an image ahead of the host vehicle, and a traveling locus of the host vehicle is estimated in the near future. A technique for determining whether or not the own vehicle deviates from the current travel lane by superimposing the track is disclosed.
[0004]
[Problems to be solved by the invention]
By the way, in the prevention of lane departure as described above, it is very important to estimate the travel locus of the host vehicle. However, when the travel locus is estimated based on the steering angle and the yaw rate, the steering is sufficiently performed when entering the curve. In addition, since there is little occurrence of yaw rate, there is a problem that the traveling curvature of the estimated traveling locus cannot be obtained with sufficient accuracy, and it is often determined that the vehicle deviates from the current traveling lane. . Further, when the S-curve or the like is turned back, the traveling curvature calculated with respect to the actual movement of the own vehicle is delayed, so that it is often determined that the own vehicle deviates from the current traveling lane.
[0005]
The present invention has been made in view of the above circumstances, can accurately determine the possibility of departure from the lane, and prevents an unnecessary increase in the case where the host vehicle is determined to depart from the current travel lane. It is an object of the present invention to provide a vehicle driving support device that is natural and easy to use.
[0006]
[Means for Solving the Problems]
In order to achieve the above object, a driving support apparatus for a vehicle according to the present invention as set forth in claim 1 comprises a travel path recognition means for recognizing a travel path ahead of the host vehicle, and a travel path estimation means for estimating the travel path of the host vehicle. In the vehicle driving support device, comprising the departure determination means for comparing the travel path and the travel locus to determine the departure of the own vehicle from the travel path, the departure determination means includes: When it is determined that the traveling state is in the state of entering the curve, the traveling path and the traveling locus are compared to determine the deviation of the host vehicle from the traveling path and set as a first determination result, The travel trajectory estimated by the travel trajectory estimation unit is corrected, the corrected travel trajectory is compared with the travel path, the deviation of the host vehicle from the travel path is determined as a second determination result, The first judgment result and the second judgment The results are characterized by determining the deviation from the running path of the vehicle.
[0007]
That is, in the vehicle driving support device according to the first aspect, the travel path recognition unit recognizes the travel path ahead of the host vehicle, the travel path estimation unit estimates the travel path of the host vehicle, and the departure determination unit determines the travel path. And the travel locus are compared to determine the deviation of the vehicle from the travel path. In this departure determination, when the departure determination means determines that the traveling state of the host vehicle is in a curve approaching state, the departure determination unit compares the traveling path with the traveling locus to determine the departure from the traveling path of the host vehicle. In addition to the first determination result, the travel trajectory estimated by the travel trajectory estimation unit is corrected, and the corrected travel trajectory is compared with the travel path to determine a deviation from the travel path of the host vehicle. As a determination result, the departure from the traveling path of the host vehicle is determined based on the first determination result and the second determination result. As described above, when it is determined that the traveling state of the host vehicle is in the state of entering the curve, the deviation from the traveling path of the host vehicle is determined based on the first determination result and the second determination result. The possibility of lane departure can be determined with high accuracy, and the case where it is determined that the vehicle departs from the current travel lane is prevented from being unnecessarily increased, resulting in a natural and usable device.
[0008]
According to a second aspect of the present invention, there is provided the vehicle driving support apparatus according to the first aspect, wherein the departure determination means is configured to output both the first determination result and the second determination result. However, when the host vehicle departs from the travel path, the host vehicle is determined to depart from the travel path, i.e., only one of the results is that the host vehicle departs from the travel path. Even in this case, this result is canceled and it is possible to prevent an unnecessary increase in the case where it is determined that the host vehicle departs from the current travel lane.
[0009]
Further, the vehicle driving support device according to the present invention as set forth in claim 3 is the vehicle driving support device according to claim 1 or 2, wherein the deviation determining means is responsive to the turning speed and the own vehicle speed. It is determined whether or not the traveling state of the host vehicle is in a curve approaching state. For example, in the case of a vehicle behavior observed on a straight road or an expressway, the traveling state of the host vehicle is in a curve approaching state. Do not judge.
[0010]
According to a fourth aspect of the present invention, there is provided the vehicle driving support apparatus according to any one of the first to third aspects, wherein the departure determination means is the second determination. The corrected travel locus used for obtaining the result is obtained by calculating the magnitude of the travel curvature estimated by the travel locus estimating means according to the vehicle speed and the steering angle, for example, entering a curve Since the earlier curve can be determined to be gentler as the speed is higher and the steering angle is smaller, correction is performed in inverse proportion to the vehicle speed and in proportion to the steering angle.
[0011]
Furthermore, the vehicle driving support apparatus according to the present invention as set forth in claim 5 is the vehicle driving support apparatus according to any one of claims 1 to 4, wherein the departure determining means is the second determination. The corrected traveling locus used for obtaining the result is obtained by a value obtained by correcting the direction of the traveling curvature estimated by the traveling locus estimating means in accordance with the turning direction, for example, the turning direction and the traveling curvature. The direction of the running curvature is corrected so that the directions of the two are the same, and erroneous determinations such as switching back at the S-curve are reduced.
[0012]
According to a sixth aspect of the present invention, there is provided the vehicle driving support apparatus according to any one of the first to fifth aspects, wherein the departure determining means is the first determination. The travel locus used for obtaining the result is obtained with a predetermined average running curvature, and the corrected running locus used for obtaining the second determination result is obtained by correcting the currently estimated running curvature. The second determination result is obtained by correcting the currently estimated traveling curvature, thereby improving the responsiveness of obtaining the traveling curvature with respect to the actual movement of the host vehicle, and quickly and accurately using an S-curve or the like. Contribute to a good judgment.
[0013]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
1 to 3 show an embodiment of the present invention, FIG. 1 is a schematic configuration diagram of a vehicle driving support device, FIG. 2 is an explanatory diagram of lane departure judgment, and FIG. 3 is a flowchart of lane departure prevention control.
[0014]
In FIG. 1, reference numeral 1 denotes a vehicle such as an automobile (own vehicle). If the vehicle 1 is extremely approached or deviated from the lane, the driver is notified by an alarm or the like. A vehicle driving support device (ADA device) 2 having a so-called lane departure prevention function as a function for preventing a traffic accident caused by carelessness such as looking away is mounted. Hereinafter, in the embodiment of the present invention, only the lane departure prevention function portion of the ADA device 2 will be described, and description of the other function portions will be omitted.
[0015]
The ADA device 2 has a pair of (left and right) CCD cameras 3 using a solid-state image pickup device such as a charge coupled device (CCD) as a stereo optical system. Attached to the front of the ceiling at a fixed interval, the object outside the vehicle is captured in stereo from different viewpoints.
[0016]
Further, the vehicle 1 is provided with a vehicle speed sensor 4 that detects the vehicle speed V, a yaw rate sensor 5 that detects the yaw rate γ, and a steering angle sensor 6 that detects the steering angle δ, and the detected vehicle speed V, The yaw rate γ, the steering angle δ, and the traveling direction image of the host vehicle 1 captured by the pair of CCD cameras 3 are input to the control device 7.
[0017]
The control device 7 receives an image of the traveling direction of the host vehicle 1, the host vehicle speed V, the yaw rate γ, and the steering angle δ, determines the possibility of lane departure ahead of the traveling path of the host vehicle 1, When it is determined that there is a possibility of lane departure, the alarm lamp 9 and the alarm buzzer 10 of the combination meter 8 are activated.
[0018]
That is, the control device 7 is formed by a multi-microprocessor system having an image processor, and a set of stereo images captured by a set of CCD cameras 3 is subjected to the principle of triangulation from a corresponding positional shift amount. A process for obtaining distance information over the entire image is performed to generate a distance image representing a three-dimensional distance distribution. Then, the distance image is processed to extract necessary information, and the possibility of departure from the lane ahead of the traveling path of the host vehicle 1 is determined based on this information.
[0019]
Specifically, a lane, a vehicle, an object other than the vehicle, and the like are detected separately from the distance image, and image data of the road and the object are obtained. With this image data, for lanes, for example, as shown in FIG. 2, the XY coordinates based on the position and orientation of the host vehicle 1 (substantially the bumper tip position is the origin O at the approximate center of the vehicle width). In the coordinate), data on the position and shape coordinates of the left and right lanes in front of the vehicle is obtained.
[0020]
On the other hand, the control device 7 uses the current vehicle speed V and the yaw rate γ, or the vehicle speed V and the steering angle δ, so that the traveling locus of the vehicle 1, that is, the turning curvature and the width dimension of the vehicle 1 (for example, 1.. 7m) is estimated. Then, a first determination line LK is set in parallel with the width direction of the host vehicle 1 at a first distance ahead of the host vehicle 1 when traveling with the estimated turning curvature (for example, about 10.24 m forward). A second determination line LE parallel to the width direction of the host vehicle is set at a second distance ahead of the host vehicle 1 (for example, approximately 18.432 m forward). Note that the position of the host vehicle 1 at about 10.24 m ahead and about 18.432 m ahead is given in advance by a function having the turning curvature as a variable. Thus, the control device 7 determines the possibility of lane departure according to the vehicle position and lane position on the first determination line LK, and determines the vehicle position and lane position on the second determination line LE. Accordingly, when the possibility of lane departure is determined and both of these two determination results indicate that there is a possibility of lane departure (when the traveling lane overlaps the estimated traveling locus of the own vehicle 1), the own vehicle 1 is finally determined as the possibility of lane departure. Here, the position of the host vehicle in the first determination line LK and the second determination line LE is such that when the left lane of the travel lane overlaps with the estimated travel locus of the host vehicle 1, the vehicle deviates to the left side or travels. When the right lane of the lane overlaps the estimated travel locus of the host vehicle 1, it is determined that the lane is deviated to the right.
[0021]
Further, the control device 7 performs the above-described lane departure determination separately depending on whether or not the traveling state of the host vehicle 1 is in a curve approach state, as will be described in detail in a flowchart of lane departure prevention control described later. It has become. Specifically, it is determined from the own vehicle speed V and the turning speed (differential value of the steering angle δ) whether or not the traveling state of the own vehicle 1 is in a curve approaching state, for example, the own vehicle speed V is 30 km / h to When it is 60 km / h and the turning speed is 40 deg / s or more, it is determined that the vehicle is entering a curve.
[0022]
If the traveling state of the host vehicle 1 does not satisfy the conditions of the host vehicle speed V and the steering speed and it can be determined that the vehicle is traveling on a straight road or a highway, the average turning curvature Cu3 for the past 0.3 seconds is determined. Based on the lane departure determination described above (first determination result).
[0023]
On the other hand, when it is determined that the traveling state of the host vehicle 1 satisfies the above-described conditions of the host vehicle speed V and the steering speed and is in the state of entering the curve, the magnitude and direction of the currently calculated turning curvature Cua are set. Correction (correction) is performed, and the above-described lane departure determination is performed based on the corrected turning curvature Cua (second determination result).
[0024]
Thus, when it is determined that the traveling state of the host vehicle 1 is not in the curve approaching state, the first determination result is determined as the final determination result, while the traveling state of the host vehicle 1 is determined to be in the curve approaching state. In this case, the first determination result and the second determination result are compared, and the final determination result is determined to be a lane departure only when both are determined to be a lane departure.
[0025]
As described above, in the present embodiment, the control device 7 has functions as travel path recognition means, travel locus estimation means, and departure determination means.
[0026]
In addition, the control device 7 issues an alarm when the host vehicle 1 has a possibility of lane departure based on the determination result of the possibility of lane departure. This warning is performed according to the vehicle speed, for example, when the vehicle speed V is 40 km / h to 50 km / h, only the warning lamp 9 is alarmed. When the vehicle speed V is 50 km / h or more, the alarm lamp 9 and the alarm buzzer 10 are used. It is controlled to perform an alarm by both.
[0027]
Next, the lane departure prevention function achieved by the above configuration will be described with reference to the lane departure prevention control flowchart shown in FIG. This lane departure prevention control program is repeatedly executed every predetermined time, for example, every 100 ms. First, in step (hereinafter abbreviated as “S”) 101, image recognition processing is executed, and as described above, a set of CCDs Based on the image of the traveling direction of the host vehicle 1 captured by the camera 3, the front left lane and the right lane are recognized.
[0028]
Thereafter, the process proceeds to S102, and necessary parameters such as the yaw rate γ, the host vehicle speed V, and the steering angle δ are read, and the process proceeds to S103.
[0029]
In S103, it is determined whether or not the yaw rate sensor 5 is valid. If the yaw rate sensor 5 is valid, the process proceeds to S104, and the current turning curvature Cua is calculated by using, for example, the following equation (1) using the yaw rate γ and the vehicle speed V. The
Cua = γ / V (1)
[0030]
On the other hand, when it is determined in S103 that the yaw rate sensor 5 is not effective, the process proceeds to S105, and it is determined whether or not the steering is performed when the steering angle δ is equal to or greater than a predetermined value (for example, 0.57 degrees).
[0031]
As a result of the determination in S105, when it can be determined that the steering angle δ is 0.57 degrees or more and the steering is being performed, the process proceeds to S106, and the following (2), (3) is performed using the steering angle δ and the own vehicle speed V. ) To calculate the current turning curvature Cua.
Re = (1 + A · V ² ) · (L / δ) (2)
Cua = 1 / Re (3)
Here, Re is a turning radius, A is a vehicle stability factor, and L is a wheelbase.
[0032]
As a result of the determination in S105, if it can be determined that the steering angle δ is less than 0.57 degrees and the steering is being performed, the process proceeds to S107, and the current turning curvature Cua is set to zero.
[0033]
After the current turning curvature Cua is set in S104, S106, or S107, the process proceeds to S108, and the average turning curvature Cu3 is calculated from the turning curvature for the past approximately 0.3 seconds including the current turning curvature Cua.
[0034]
Thereafter, the process proceeds to S109, where the vehicle positions of the first judgment line (10.24m) and the second judgment line (18.432m) are estimated based on the average turning curvature Cu3 and deviated by comparison with the white line position. A determination is made, and this determination result is taken as a first determination result. That is, the lane departure is determined only when the lane overlaps the traveling locus of the host vehicle 1 in both the first determination line and the second determination line.
[0035]
Next, in S110, whether or not the traveling state of the host vehicle 1 is in a curve approach state, that is, when the host vehicle speed V is 30 km / h to 60 km / h and the turning speed is 40 deg / s or more. If the travel state of the host vehicle 1 does not satisfy the conditions of the host vehicle speed V and the steering speed, and it can be determined that the vehicle is traveling on a straight road or a highway, the process proceeds to S111. The first determination result is substituted into the determination result, and this is used as the second determination result.
[0036]
On the other hand, if it is determined in S110 that the traveling state of the host vehicle 1 satisfies the above-described conditions of the host vehicle speed V and the turning speed and the vehicle is in a curve approaching state, the process proceeds to S112 and the current turning curvature Cua is increased. Correct the size. Specifically, the correction coefficient k to be multiplied by the current turning curvature Cua is set as shown in the following equation (4).
[0037]
k = | α · (δ / V) | (4)
Here, α is a gain. That is, for example, as the approach speed to the curve is high and the steering angle is small, it can be determined that the previous curve is gentle. Therefore, the correction is performed in inverse proportion to the vehicle speed V and in proportion to the steering angle δ. The gain α is prepared for the first determination line (10.24 m) and the second determination line (18.432 m), and the gain α is set to a small value on the first determination line (10.24 m). In the second judgment line (18.432 m), the value is set to a large value, and the running curvature is corrected in accordance with the curved curve. By correcting the current turning curvature Cua in this way, the estimated current turning curvature Cua is accurate even if the steering angle δ and the yaw rate γ are small values when the host vehicle 1 enters the curve. It is corrected to the value.
[0038]
Then, it progresses to S113 and corrects the direction of the present turning curvature Cua. Specifically, the direction of the current turning curvature Cua is compared with the steering direction, and if they do not match, the correction coefficient k by which the current turning curvature Cua is multiplied is set to k = −k, and the direction matches the steering direction. Correct as follows. That is, the yaw rate γ is corrected in consideration of the case where the yaw rate γ does not sufficiently appear in the steering direction due to the turning-back at the S-curve.
[0039]
Next, the process proceeds to S114, where the current turning curvature Cua is multiplied by the correction coefficient k corrected (corrected) in S112 and S113 described above, and the first judgment line (10.24 m) is based on the corrected current turning curvature Cua. ), The vehicle position of the second determination line (18.432 m) is estimated, a departure determination is performed by comparison with the position of the white line, and this determination result is used as the second determination result. That is, the lane departure is determined only when the lane overlaps the traveling locus of the host vehicle 1 in both the first determination line and the second determination line. Thus, since the second determination result is obtained based on the corrected current turning curvature Cua, an accurate determination result with a good response can be obtained even on a traveling road in which left and right curves are continuous.
[0040]
After obtaining the second determination result in S111 or S114, the process proceeds to S115, where the first determination result and the second determination result are compared. As a result of this comparison, if the two deviation determination results are both deviated, the process proceeds to S116 and a deviant alarm is generated. Otherwise, the process proceeds to S117, and the routine is exited without performing the deviant alarm.
[0041]
Then, as described above, the departure warning performed in S116 is performed according to the vehicle speed, for example, when the vehicle speed V is 40 km / h to 50 km / h, only the warning lamp 9 is alarmed, and the lane departure to the left is performed. Whether the lane is deviating to the right side is also notified by turning on the warning lamp 9. Further, when the vehicle speed V is 50 km / h or more, an alarm is given by the alarm buzzer 10 in addition to the alarm lamp 9.
[0042]
Thus, according to the present embodiment, when it is determined that the traveling state of the host vehicle 1 is in the state of entering the curve, the lane and the traveling locus are compared to determine the departure from the lane of the host vehicle 1. The first determination result is corrected, the estimated current travel locus is corrected, the corrected current travel locus is compared with the lane, and a deviation from the lane of the host vehicle 1 is determined to determine the second determination result. A departure from the lane of the host vehicle is determined based on the first determination result and the second determination result. That is, as shown in FIG. 2, when it is determined that the traveling state of the host vehicle 1 is in a curve approaching state, the traveling locus conventionally given by Sk1 in the drawing corrects the current traveling locus. And given by Sk2. For this reason, in the conventional travel locus Sk1, the point PKL0 on the lane is located on the inner side of the point PKL1 on the travel locus on the first determination line LK, and on the inner side of the point PEL1 on the travel locus on the second determination line LE. The point PEL0 on the lane is positioned at the wrong position, and it is erroneously determined as a lane departure. On the other hand, in the present embodiment, the point PKL0 on the lane is located outside the point PKL2 on the travel locus on the first determination line LK, and outside the point PEL2 on the travel locus on the second determination line LE. If the point PEL0 on the lane is located and does not depart from the lane, the second determination result can be used for accurate determination. As a result, in this embodiment, it is finally determined that the vehicle does not depart from the lane. Even when the yaw rate γ or the steering angle δ is small, such as when entering a curve, an accurate lane departure can be determined. Therefore, it is possible to prevent an unnecessary increase in the number of cases determined to deviate from the above, and it is a natural and easy-to-use device.
[0043]
【The invention's effect】
As described above, according to the present invention, it is possible to accurately determine the possibility of a lane departure, preventing an unnecessary increase in the case where the host vehicle is determined to depart from the current travel lane, The effect of being natural and easy to use can be obtained.
[Brief description of the drawings]
FIG. 1 is a schematic configuration diagram of a vehicle driving support device. FIG. 2 is an explanatory diagram of lane departure judgment. FIG. 3 is a flowchart of lane departure prevention control.
DESCRIPTION OF SYMBOLS 1 Own vehicle 2 Driving support apparatus 3 for vehicles CCD camera 7 Control apparatus (travel path recognition means, travel locus estimation means, deviation determination means)

Claims

Traveling path recognition means for recognizing a traveling path ahead of the host vehicle;
Traveling locus estimating means for estimating the traveling locus of the host vehicle;
In a vehicle driving support device comprising a departure determination means for comparing the travel path and the travel locus to determine a deviation of the host vehicle from the travel path,
The departure determining means, when determining that the traveling state of the host vehicle is in a curve approaching state, compares the traveling path with the traveling locus to determine the departure of the host vehicle from the traveling path. The travel locus estimated by the travel locus estimation means is corrected and the corrected travel locus is compared with the travel route to determine the departure of the host vehicle from the travel route. A vehicle driving support apparatus characterized in that a deviation from the travel path of the host vehicle is determined based on the first determination result and the second determination result as a second determination result.

The departure determination means determines that both the first determination result and the second determination result determine that the own vehicle deviates from the traveling road when the own vehicle departs from the traveling road. The vehicle driving support device according to claim 1.

The said deviation determination means determines whether the driving state of the said own vehicle exists in the state of curve approach according to the speed of steering and the own vehicle speed, The claim 1 or Claim 2 characterized by the above-mentioned. Vehicle driving support device.

The deviation determination unit uses the corrected travel locus used for obtaining the second determination result as a value obtained by correcting the magnitude of the traveling curvature estimated by the travel locus estimation unit according to the vehicle speed and the steering angle. The vehicle driving support device according to any one of claims 1 to 3, wherein the driving assistance device for a vehicle is obtained.

The deviation determination means obtains the corrected traveling locus used for obtaining the second determination result by a value obtained by correcting the direction of the traveling curvature estimated by the traveling locus estimation means according to the turning direction. The driving support device for a vehicle according to any one of claims 1 to 4, wherein the driving support device is for a vehicle.

The deviation determination means obtains the traveling locus used for obtaining the first determination result with a predetermined average traveling curvature, and the corrected traveling locus used for obtaining the second determination result is currently estimated. 6. The vehicle driving support apparatus according to claim 1, wherein the driving curvature is obtained by correcting the traveling curvature.