JP4082471B2 - Outside monitoring device - Google Patents

Description

【００８０】
また、各頂点の旋回半径ｒnは、以下の(12)式で求めることができるため、これを用い、旋回半径が最大の（最も外側を通る）頂点番号ｎoと最小の（最も内側を通る）頂点番号ｎiを求める。そして、頂点番号ｎoの現在の座標Ｐｎoと予測時間ｔ後の座標Ｐ'ｎoとを、(12)式を適用して求めた最大旋回半径ｒnoの円弧で繋ぐとともに、頂点番号ｎiの現在の座標Ｐｎiと予測時間ｔ後の座標Ｐ'ｎiとを、(12)式を適用して求めた最小旋回半径ｒniの円弧で繋ぎ、残った頂点を、それぞれ隣り合う頂点と直線で繋ぐことにより、図１６に示すように、自車両の通過軌跡を得ることができる。
ｒn＝((Ｐxn−Ｏx)²＋(Ｐzn−Ｏz)²)^1/2 …(12)
【００８１】
この場合、舵角０すなわち旋回半径無限大のときには、計算実行が不可能になるため、予測時間ｔ後の頂点ｎの座標は、以下の(13)式で求め、左右それぞれに最も張り出した頂点を直線で繋いで推定通過軌跡とする。

【００８２】
そして、自車両の通過軌跡を推定した後、上記ステップS201からステップS202へ進み、後述する接触判定及び隙間計測のため、自車両の輪郭モデル（図１５参照）の先端中央Ｃｆ＝[Ｃxf,Ｃzf]^Tと後端中央Ｃｒ＝[Ｃxr,Ｃzr]^Tとを、各頂点と同様に予測時間ｔ後の位置Ｃ'ｆ，Ｃ'ｒを算出し、図１７に示すように、通過軌跡による走行領域を右側の判定領域と左側の判定領域とに分割する。
【００８３】
続くステップS203では、先の立体物検出処理で得られた各グループの速度Ｖｘ，Ｖｚから、予測時間ｔにおける立体物のＸ方向移動量ｄｘ、Ｚ方向移動量ｄｚを以下の(14),(15)式によって求め、各距離データをＸ方向移動量ｄｘ、Ｚ方向移動量ｄｚだけ移動させ、予測時間ｔ後の立体物の位置を推定する。
ｄｘ＝Ｖｘ・ｔ …(14)
ｄｚ＝Ｖｚ・ｔ …(15)
【００８４】
図１３に示す立体物の位置の検出結果をグループ化した例では、植え込みを検出したグループ１，２，３、及び、右の壁を検出したグループ７ではＸ方向、Ｚ方向ともに速度は略０と検出されるため、予測時間ｔ後の位置は検出位置から変化しない。また、先行車両を検出したグループ４では、例えば、速度Ｖｘ＝０、Ｖｚ＝４．０ｍ／ｓｅｃと検出されたものとし、ｔ＝１．５ｓｅｃと仮定すると、グループ４の移動量はｄｘ＝０．０ｍ、ｄｚ＝６．０ｍとなる。また、自転車を検出したグループ６は、例えば、Ｖｘ＝−０．１ｍ／ｓｅｃ、Ｖｚ＝−１．０ｍ／ｓｅｃと検出されたとすると、移動量は、ｄｘ＝−０．１５ｍ、ｄｚ＝−１．５ｍとなる。図１８は、各距離データを、Ｘ方向移動量ｄｘ、Ｚ方向移動量ｄｚだけ移動させた予測時間ｔ後の推定位置を示している。
【００８５】
その後、ステップS204へ進み、ステレオカメラの視野外となった立体物について、距離データの追跡を行う。Ｘ方向及びＺ方向とも速度略０と検出された距離データのグループは道路上に静止している立体物とし、その距離データが各処理サイクル毎に保持されている。
【００８６】
ｋ番目の処理サイクルとｋ−１番目の処理サイクル間の走行軌跡は、車速センサ４及び舵角センサ５により、走行軌跡Ｌ'、旋回半径Ｒ'として得ることができる。このとき、走行距離Ｌ'は、ｋ番目の処理サイクルでの位置を基準にするため、Ｌ'＜０となり、旋回中心の座標をＯ'＝[Ｏ'x,Ｏ'z]^Tとし、旋回角度をαとすると、この旋回角度αは、以下の(16)式によって与えられる。
α＝Ｌ'／Ｒ' …(16)
【００８７】
従って、ｋ−１番目の処理サイクルでの距離データをＤ＝[Ｄx,Ｄz]^Tとすると、ｋ番目の処理サイクルでの距離データＤ'は、以下の(17)式によって求めることができ、これを処理サイクル毎に更新してゆくことにより、図１９に示すように、ステレオカメラの視野外に逸脱してしまった距離データを追跡することができる。

【００８８】
その後、上記視野外に逸脱してしまった立体物の追跡データを含め、自車両通過軌跡と立体物の接触判定を行うため、ステップS205で、予測時間ｔ後の位置を推定した距離データの中から最初のデータを読み込むと、ステップS206で、その推定距離データが右側判定領域内にあるか否かを調べる。
【００８９】
そして、推定距離データが右側判定領域内にない場合には、上記ステップS206からステップS208へ進み、右側判定領域内にある場合、自車両と接触する可能性があると判断して上記ステップS206からステップS207へ進んで右側の接触を示す右側接触フラグをＯＮし、ステップS208へ進む。
【００９０】
ステップS208では、推定距離データが左側判定領域内にあるか否かを調べ、左側判定領域内にない場合、ステップS210へ進み、推定距離データが左側判定領域内にある場合、ステップS211で左側の接触を示す左側接触フラグをＯＮし、ステップS210へ進む。
【００９１】
ステップS210では、右側あるいは左側の接触フラグがＯＮされたか否かを調べ、いずれかの側の接触フラグがＯＮされたときには、ステップS211で警報を発生し、ディスプレイ９への警告表示、あるいは、ランプ、ブザー等による警告を行い、運転者に知らせる。
【００９２】
一方、上記ステップS210で、左右の接触フラグがＯＮされていないときには、上記ステップS210からステップS212へ進み、接触判定領域外の推定距離データを、通過軌跡前面より前方にあるデータを除いて車体中心線左右にグループに振り分ける。次に、ステップS213で、接触判定領域外の推定距離データと自車両の通過軌跡との間の隙間距離を算出する。この隙間距離は、図２０に示すように、左グループの場合は自車両左側面の通過軌跡に対して、右グループの場合は自車両右側面の通過軌跡に対して法線を引き、この長さを“隙間距離”とし、その値を記憶する。
【００９３】
その後、ステップS214へ進んで最終の推定距離データか否かを調べ、最終の推定距離データに達していないときには、ステップS215で次の推定距離データを読み込んで上記ステップS206へ戻り、以上の処理を繰り返す。そして全ての推定距離データについて接触判定を完了すると、上記ステップS214からステップS216へ進み、左右各グループ毎に隙間距離の最小値を算出して保持し、ステップS217で、数値表示あるいは算出した隙間距離の最小値に応じて注意を喚起するための表示信号をディスプレイ９に出力し、運転者に報知する。
【００９４】
図２１は、図１８に示す予測時間ｔ後の各距離データの推定位置に対し、自車両の走行領域を記入したものであり、この例では、先行車両の現在の検出位置は自車両の走行領域に入っているが、立体物の速度と自車両の走行軌跡とを考慮し、設定時間後の推定位置に対して接触判定を行うため、先行車両とは接触しないと正しく判定することができる。
【００９５】
また、図２１における自転車のように複雑な対象物に対しても、グループ内の個々の距離データをグループの速度に応じて移動させるため、検出される対象物の凹凸がそのまま保持され、接触判定を精密に行うことができる。図２１の例では、自転車に接触する危険があると正しく判定され、思わぬ事故を未然に回避することができる。図２１においては、植え込みのように、静止している立体物に対しても同様である。
【００９６】
しかも、この場合、自車両の輪郭が描く走行軌跡を、自車両の速度及び舵角に応じて推定するため、対象物の凹凸を考慮することに合わせ、自車両の凹凸も考慮され、接触判定、隙間計測の精度を向上することができるばかりでなく、運転者の注視する領域と接触判定領域とを近づけることができ、違和感の少ない警報を発することができる。
【００９７】
さらに、検出範囲外に逸脱してしまった対象物に対しても距離データの追跡を行うため、自車両のすぐ横の立体物に対しても、接触判定、隙間計測を行うことができ、特に、内輪差の影響による接触事故防止に大きな効果を発揮することができる。
【００９８】
図２２及び図２３は本発明の実施の第２形態に係わり、図２２は立体物検出処理の部分フローチャート、図２３はグループの対応関係を示す説明図である。
【００９９】
前述の第１形態では、立体物の移動速度を検出するため、検出したグループを“物体”又は“側壁”として直線状に近似し、その中心点を求める処理を行っている。これに対し、本形態では、前回処理で検出したグループ内の個々の距離データと今回処理で検出したグループ内の個々の距離データとを比較し、グループとしての移動量を求め、速度を算出するものである。
【０１００】
このため、本形態では、画像処理用コンピュータ３０のマイクロプロセッサ３０ｂによる立体物検出処理の一部を変更する。本形態の立体物検出処理は、第１形態の図３〜図６のプログラムにおいて、ステップS101〜S114の距離データ検出処理、ステップS115〜S121の距離グループ検出処理、ステップS122〜S132のグループ分割処理までは同じであり、このグループ分割処理以降を変更する。
【０１０１】
すなわち、本形態では、ステップS122〜S132のグループ分割処理を完了すると、ステップS132から図２２のステップS170へ進み、ステップS170〜S174で、前回処理で検出された複数のグループの中から今回の処理で検出されたグループと、同一の立体物を検出しているグループを検索してグループの対応関係を求める処理を行った後、ステップS175以降で、今回処理で検出されたグループと対応関係にある前回検出のグループとの間の移動量からグループの速度を求める。
【０１０２】
このため、まず、ステップS170で今回検出の最初のグループのデータを読み込むと、ステップS171で、グループ内の距離データの重心点（Ｘｃ，Ｚｃ）を求める。この重心点（Ｘｃ，Ｚｃ）は、グループ内のｉ番目の距離データの位置をｘi、ｚiとし、ｎをグループに所属する距離データの個数とすると、以下の(18),(19)式によって求めることができる。但し、Σはｉ＝１〜ｎの総和を求めるものとする。
Ｘｃ＝（Σｘi）／ｎ …(18)
Ｚｃ＝（Σｚi）／ｎ …(19)
【０１０３】
次に、ステップS172へ進み、前回処理で検出されたグループの重心点を読み出し、この前回検出のグループの重心点と今回検出のグループの重心点との間の距離（重心点の移動量）を算出すると、ステップS173で、各グループの重心点間の距離が判定値以下か否かを調べる。そして、前回処理で検出されたグループの重心点と今回検出のグループの重心点との間の距離が判定値を超えているときには、これらのグループは同一立体物を検出したものではないと判断してステップS174へ進み、前回検出の次のグループの重心点を読み出して上記ステップS172へ戻る。
【０１０４】
一方、上記ステップS173で、各グループの重心点間の距離が判定値以下のときには、これらのグループは同一立体物を検出した対応関係にあると判断して上記ステップS173からステップS175へ進み、前回検出のグループ内の各距離データを移動させてグループ間のズレ量が最小となる修正量を算出する。
【０１０５】
例えば、図２３に示すように、前回検出のグループＦと今回検出のグループＧが対応関係にあるとすると、まず、前回検出のグループＦの各距離データを、今回検出のグループＧの重心点ＰGと前回検出のグループＦの重心点ＰFとが一致するように移動する。
【０１０６】
次に、この移動後のグループＦ１の各距離データｉF1について、今回検出のグループＧ内で最も距離が近い距離データｉGを求め、各距離データｉF1,ｉG間の距離ｄｉF1を求める。これを移動後のグループＦ１内の全ての距離データについて行い、距離ｄｉF1の二乗和ＳＤＫ1（ズレ量）を、以下の(20)式によって求める。
ＳＤＫ1＝ΣｄｉF1² …(20)
【０１０７】
さらに、グループＦ１の全体の位置を、Ｚ方向及びＸ方向に微小量Δｚ，Δｘだけ移動し、この移動後のグループＦ２と今回検出のグループＧとの間のズレ量ＳＤＫ2を、上記と同様に算出する。そして、この処理を、Δｚ，Δｘを変化させて繰り返し、各ずれ量ＳＤＫ1,ＳＤＫ2,ＳＤＫ3,…の中で最小値が得られる微小量Δｚ，Δｘの値を、グループ間のズレ量が最小となる修正量とする。
【０１０８】
以上により、グループ間のズレ量が最小となる修正量を求めると、次に、ステップS176へ進み、対応関係にあるグループ間の重心点の移動量に上記修正量を加えた値をグループ間の移動量とし、ステップS177で、この移動量を処理サイクル毎の時間間隔で割り算してグループの速度Ｖｚ，Ｖｘを算出する。次いで、ステップS178で最終グループか否かを調べ、最終グループでないときには、ステップS179で次のグループのデータを読み込んでステップS171へ戻り、以上の処理を繰り返す。尚、簡易的には、上記ステップS172で算出した重心点の移動量を処理サイクル毎の時間間隔で割り算し、グループの速度としても良い。
【０１０９】
そして、上記ステップS178で最終グループとなって最終グループの処理が完了したとき、上記ステップS178からステップS180へ進んで各グループの位置や速度等のパラメータをメモリに書き込み、立体物検出処理の全体のプログラムを終了する。この立体部検出処理のデータを用いる以降の隙間距離算出処理は、前述の第１形態と同じである。
【０１１０】
本形態では、前述の第１形態に対し、立体物の移動速度を高い精度で検出することができ、接触判定や隙間距離の精度をより一層向上することができる。
【０１１１】
図２４〜図２７は本発明の第３形態に係り、図２４は車外監視装置の全体構成図、図２５は車外監視装置の回路ブロック図、図２６は画像の例を示す説明図、図２７は立体物の二次元分布の例を示す説明図である。
【０１１２】
本形態の車外監視装置５０は、図２４に示すように、上下２台のカメラからなるステレオ光学系６０からの画像を処理して立体物の二次元的な位置分布を認識する通路パターン認識装置７０を採用し、この通路パターン認識装置７０に第１形態と同様の画像処理用コンピュータ３０を接続したものであり、通路パターン認識装置７０によって得られる立体物の二次元分布の位置情報を前述の第１形態と同様の方法で処理するものである。
【０１１３】
上記通路パターン認識装置７０は、周知の装置を適用することができ、例えば、計測自動制御学会論文集Ｖｏｌ．２１，Ｎｏ．２（昭和６０年２月）「障害物の２次元的な分布の認識手法」に記載されている通路パターン認識装置等を適用することができる。尚、上記通路パターン認識装置７０は、前述の第１形態と同様に２台のカメラの画像を処理して被写体までの距離分布を検出するものであるが、装置内部のデータ処理方法が異なるため立体物の情報のみが出力され、第１形態のような白線による道路形状の検出はできない。
【０１１４】
上記ステレオ光学系６０は、車両４０の前部に上下一定の間隔をもって取り付けられる２台のＣＣＤカメラ６０ａ，６０ｂで構成されており、また、上記通路パターン認識装置７０は、図２５に示すように、上記ステレオ光学系６０からの上下２枚の画像信号を入力し、立体物の距離及び位置の二次元分布を算出する距離検出回路６０ａ、この距離検出回路６０ａからの二次元分布情報を記憶する二次元分布メモリ６０ｂ等から構成されている。
【０１１５】
本形態では、ステレオ光学系６０で撮像した画像、例えば、図２６に示すような画像を通路パターン認識装置７０で処理すると、図２７に示すような立体物の二次元分布パターンが出力される。これは、前述の第１形態における区分毎の距離データに相当するものであり、この立体物の二次元分布パターンに対し、前述の第１形態と同様の処理を行うことにより、立体物検出、接触判定、隙間距離計測を行うことができる。
【０１１６】
本形態では、側壁の有無や位置を、直線式のパラメータや前後端の座標といった簡素なデータ形態に変換するため、利用側でのデータの取扱いや処理が容易となる。
【０１１７】
図２８〜図３２は本発明の実施の第４形態に係わり、図２８は車外監視装置の全体構成図、図２９は車外監視装置の回路ブロック図、図３０はレーザビームの走査方法を側面から示す説明図、図３１はレーザビームの走査方法を上面から示す説明図、図３２はレーザレーダ測距装置で計測される立体物の二次元分布の例を示す説明図である。
【０１１８】
前述の各形態ではカメラの撮像画像を処理して車外の物体を検出するようにしているが、これに対し、本形態は、レーザビームの走査によって車外の物体を検出するものである。すなわち、図２８に示すように、本形態の車両８０に搭載される車外監視装置９０は、レーザビームによるレーザレーダ測距装置１００を採用し、このレーザレーダ測距装置１００に画像処理用コンピュータ３０を接続したものである。
【０１１９】
上記レーザレーダ測距装置１００は、レーザビームを投射し、このレーザビームが物体に当たって反射してくる光を受光し、この所要時間から物体までの距離を測定するものであり、本形態の車外監視装置９０には周知のレーザレーダ測距装置を適用することができる。
【０１２０】
本形態の車外監視装置９０では、レーザビームの投射・受光と左右方向への走査機能を有するレーザ投光ユニット１０１が車両の前部に取り付けられており、図２９に示すように、レーザレーダ測距装置１００には、レーザービームの投光受光の所要時間から物体までの距離を計算し、また、レーザビームを走査する方向から物体の二次元の位置を計算する距離検出回路１００ａ、検出された物体の二次元の位置を書き込む二次元分布メモリ１００ｂ等から構成されている。
【０１２１】
図３０に示すように、レーザ投光ユニット１０１からはレーザビームが水平に投射され、道路表面より高い位置にある立体物のみが検出される。また、図３１に示すように、レーザビームは左右方向に走査され、所定の走査範囲で一定の間隔毎にレーザビームが投光・受光されて距離を検出する動作が繰り返され、立体物の二次元分布が計測される。
【０１２２】
例えば、前方左側にガードレール、右前方に他の車両がある状況を上記レーザレーダ測距装置１００で計測すると、図３２に示すような立体物の二次元分布の情報が得られる。これは、前述の第１形態における区分毎の立体物の距離データと同様である。
【０１２３】
従って、レーザレーダ測距装置１００の出力である立体物の二次元分布に対し、第１形態と同様の処理を行なうことにより、物体や壁面を検出することができる。本形態では、立体物のデータを処理が容易な形態で得ることができ、計算処理量を低減することが可能である。
【０１２４】
【発明の効果】
以上説明したように本発明によれば、検出した障害物の位置の変化や細部形状に関する情報を取り入れて自車両との接触可能性を正確且つ精密に判断することができる等優れた効果が得られる。
【図面の簡単な説明】
【図１】本発明の実施の第１形態に係わり、車外監視装置の全体構成図
【図２】同上、車外監視装置の回路ブロック図
【図３】同上、立体物検出処理のフローチャート（その１）
【図４】同上、立体物検出処理のフローチャート（その２）
【図５】同上、立体物検出処理のフローチャート（その３）
【図６】同上、立体物検出処理のフローチャート（その４）
【図７】同上、隙間距離算出処理のフローチャート
【図８】同上、車載のカメラで撮像した画像の例を示す説明図
【図９】同上、距離画像の例を示す説明図
【図１０】同上、距離画像の区分を示す説明図
【図１１】同上、検出対象の状況例を示す説明図
【図１２】同上、区分毎の立体物の検出距離の例を示す説明図
【図１３】同上、区分毎の立体物の検出距離をグループ化した例を示す説明図
【図１４】同上、検出した物体と側壁の結果を示す説明図
【図１５】同上、自車両輪郭の近似を示す説明図
【図１６】同上、自車両の通過軌跡を示す説明図
【図１７】同上、左右判定領域の説明図
【図１８】同上、設定時間後の立体物の位置の推定結果を示す説明図
【図１９】同上、視野外距離データの追跡を示す説明図
【図２０】同上、通過軌跡と距離データとの隙間計測の説明図
【図２１】同上、自車両と立体物との接触判定の説明図
【図２２】本発明の実施の第２形態に係わり、立体物検出処理の部分フローチャート
【図２３】同上、グループの対応関係を示す説明図
【図２４】本発明の第３形態に係り、車外監視装置の全体構成図、
【図２５】同上、車外監視装置の回路ブロック図
【図２６】同上、画像の例を示す説明図
【図２７】同上、立体物の二次元分布の例を示す説明図
【図２８】本発明の実施の第４形態に係わり、車外監視装置の全体構成図
【図２９】同上、車外監視装置の回路ブロック図
【図３０】同上、レーザビームの走査方法を側面から示す説明図
【図３１】同上、レーザビームの走査方法を上面から示す説明図
【図３２】同上、レーザレーダ測距装置で計測される立体物の二次元分布の例を示す説明図
【符号の説明】
１ …車外監視装置
１０ａ，１０ｂ…ＣＣＤカメラ
２０…イメージプロセッサ
３０…画像処理用コンピュータ[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an out-of-vehicle monitoring device that recognizes a situation outside a vehicle based on a pseudo image indicating a distance distribution of an object outside the vehicle, and in particular, detects a gap between the host vehicle and an obstacle when passing through a narrow road. The present invention relates to a vehicle exterior monitoring device that secures
[0002]
[Prior art]
Conventionally, in vehicles such as automobiles, for example, disclosed in Japanese Utility Model Laid-Open No. 5-68742 as a supplement to the driver's operational feeling when passing through narrow roads where a large number of eaves, guardrails, utility poles, parked vehicles, etc. exist. A tactile sensor such as a corner pole or a contact type switch that turns on when a bar-shaped member touches an obstacle may be attached to the vehicle body. Can be confirmed.
[0003]
In addition, recently, ultrasonic sensors are attached to the sides and four corners of a vehicle, the distance is measured by emitting ultrasonic waves and receiving reflected waves from obstacles, and informs the driver of the measured distances to narrow roads. Techniques have been developed to reduce the burden of passing through the road.
[0004]
However, in the case of attaching a mark such as the aforementioned corner pole outside the vehicle body, the driver's habituation is required, so the driver's burden reduction effect is small. Further, in the case of a contact type sensor such as a tactile sensor, the position cannot be confirmed until it comes into contact with an object, and there is a situation in which the handle operation cannot be performed in time after touching an obstacle.
[0005]
In addition, in the technology using ultrasonic waves, the spatial resolution is inferior, so it is not only possible to know the positional relationship of obstacles, but also the object dependence that the emitted ultrasonic waves do not return from pedestrian clothes or smooth walls. It is difficult to deal with various three-dimensional objects existing on the road.
[0006]
In order to deal with this, the present applicant, in Japanese Patent Application Laid-Open No. Hei 7-192199, takes a stereo image of an object outside the vehicle, and a distance distribution over the entire image by the principle of triangulation from the deviation amount of the corresponding position of the captured stereo image pair The three-dimensional position of each part of the subject corresponding to the distance distribution information is calculated, and a plurality of three-dimensional objects are detected using the information of these three-dimensional positions. A technique is proposed in which the closest distance between the vehicle side edge and the extension line of the host vehicle side is calculated as a gap distance to the left and right respectively, and information related to the left and right gap distances is notified to the driver.
[0007]
[Problems to be solved by the invention]
In the technique previously proposed by the present applicant, obstacles such as detected vehicles and guardrails are approximated by contour lines and straight lines, and a gap between them is calculated. Therefore, it is necessary to give a certain degree of margin to the gap distance in consideration of a vehicle that may be ignored in the process of approximation processing, a protrusion on the side surface of the guard rail, and the like.
[0008]
Further, in the previous technology, the gap distance is calculated on the assumption of the position at the time when the obstacle is detected. For example, the vehicle is passing through a narrow road with a preceding vehicle in front of the host vehicle. In the situation, a gap between the preceding vehicle and the host vehicle is detected. Therefore, in a situation where the preceding vehicle is standing in front as an obstacle, it is difficult to detect the gap distance between the left and right obstacles.
[0009]
The present invention has been made in view of the above circumstances, and includes an out-of-vehicle monitoring device capable of accurately and accurately determining the possibility of contact with the host vehicle by taking in information on a change in the position of a detected obstacle and a detailed shape. It is intended to provide.
[0010]
[Means for Solving the Problems]
According to the first aspect of the present invention, the pseudo image of the distance distribution obtained by measuring the distance of the object outside the vehicle is divided into a plurality of sections, and the presence / absence of the three-dimensional object and the distance data are detected for each section to detect the outside of the vehicle. In the out-of-vehicle monitoring device for recognizing the situation, means for collecting the adjacent sections in which the distance data is close to each other into one group and calculating the moving speed of the three-dimensional object for each group; Of all categories belonging to Distance data The category belongs Means for estimating the position of the three-dimensional object after the set time, moving according to the moving speed of the group, means for estimating the traveling locus drawn by the contour of the host vehicle after the set time, and the estimated position of the three-dimensional object Based on the means for determining whether or not the own vehicle is in contact with the three-dimensional object, the estimated position of the three-dimensional object and the running locus of the own vehicle, And a means for calculating a minimum gap between the three-dimensional object.
[0011]
The invention according to claim 2 is the invention according to claim 1, wherein each distance data in the group detected in the previous process is compared with each distance data in the group detected in the current process. And the movement speed is calculated from the temporal movement amount between the corresponding groups.
[0012]
The invention according to claim 3 is the invention according to claim 1, wherein the distance data of the three-dimensional object stationary on the road is held for each processing cycle, and when the three-dimensional object deviates from the detection range, A means for estimating the position of the three-dimensional object based on the distance data and the traveling locus of the host vehicle is provided.
[0013]
According to a fourth aspect of the present invention, in the first aspect of the invention, the pseudo image is a three-dimensional image obtained by calculating a correlation between a pair of images taken by a pair of cameras and using a principle of triangulation from parallax with respect to the same object. It is characterized by showing the distance distribution. The invention according to claim 5 is the invention according to claim 1, wherein the pseudo image shows a two-dimensional distance distribution obtained by processing a pair of images captured by a set of cameras. And
[0014]
The invention described in claim 6 is the invention described in claim 1, wherein the pseudo image shows a two-dimensional distance distribution obtained by projecting / receiving light of a laser beam or transmitting / receiving a radio wave. To do.
According to a seventh aspect of the present invention, in the out-of-vehicle monitoring device that measures the distance of an object outside the vehicle and detects the presence / absence of a three-dimensional object and the distance data for the three-dimensional object to recognize the situation outside the vehicle, the distance data are adjacent to each other. And the approaching objects are grouped into one group, and the moving speed of the three-dimensional object for each group is calculated. Of all categories belonging to Distance data The category belongs Means for estimating the position of the three-dimensional object after the set time, moving according to the moving speed of the group, means for estimating the traveling locus drawn by the contour of the host vehicle after the set time, and the estimated position of the three-dimensional object And means for calculating a clearance distance between the host vehicle and the three-dimensional object.
[0015]
That is, in the vehicle exterior monitoring device of the present invention, the pseudo image of the distance distribution obtained by measuring the distance to the object outside the vehicle is divided into a plurality of sections, and the presence / absence of a three-dimensional object and distance data are detected for each section. Later, when the distance data approaches each other and the adjacent sections are grouped into one group and the movement speed of the three-dimensional object for each group is calculated, Of all categories belonging to Distance data The category belongs It moves according to the moving speed of the group, estimates the position of the three-dimensional object after the set time, and estimates the travel locus drawn by the contour of the host vehicle after the set time. Then, the estimated position of the three-dimensional object is compared with the traveling locus of the own vehicle to determine whether or not the own vehicle is in contact with the three-dimensional object, and based on the estimated position of the three-dimensional object and the traveling locus of the own vehicle. The minimum gap between the host vehicle and the three-dimensional object is calculated.
[0016]
In this case, the moving speed of the three-dimensional object is determined by comparing each distance data in the group detected in the previous process with each distance data in the group detected in the current process, and checking the correspondence between the groups. It is desirable to calculate from the amount of temporal movement between corresponding groups. In addition, for a three-dimensional object that is stationary on the road, the distance data is held for each processing cycle, and when the three-dimensional object deviates outside the detection range, the previous distance data and the travel of the host vehicle are stored. It is desirable to estimate the position of the three-dimensional object based on the locus.
[0017]
In addition, the above processing is a pseudo process showing a three-dimensional distance distribution obtained by processing a pair of images captured by a pair of cameras, obtaining a correlation between the captured images, and calculating a triangulation principle from parallax with respect to the same object. A pseudo image showing a two-dimensional distance distribution obtained by processing an image, a pair of images captured by a pair of cameras, and a two-dimensional distance distribution by projecting / receiving a laser beam or transmitting / receiving a radio wave This can be performed based on the obtained pseudo image.
Furthermore, in another vehicle exterior monitoring device according to the present invention, after measuring the distance to an object outside the vehicle and detecting the presence / absence of a three-dimensional object and the distance data for the three-dimensional object, the distance data adjacent to and approaching one another When grouping and calculating the movement speed of the three-dimensional object for each group, Of all categories belonging to Distance data The category belongs It moves according to the moving speed of the group, estimates the position of the three-dimensional object after the set time, and estimates the travel locus drawn by the contour of the host vehicle after the set time. Then, the estimated position of the three-dimensional object is compared with the travel locus of the host vehicle, and the gap distance between the host vehicle and the three-dimensional object is calculated.
[0018]
DETAILED DESCRIPTION OF THE INVENTION
Embodiments of the present invention will be described below with reference to the drawings. 1 to 21 relate to a first embodiment of the present invention, FIG. 1 is a general configuration diagram of the outside monitoring apparatus, FIG. 2 is a circuit block diagram of the outside monitoring apparatus, and FIGS. 3 to 6 are three-dimensional object detection processes. FIG. 7 is a flowchart of a gap distance calculation process, FIG. 8 is an explanatory diagram showing an example of an image captured by an in-vehicle camera, FIG. 9 is an explanatory diagram showing an example of a distance image, and FIG. FIG. 11 is an explanatory diagram showing an example of the situation of the detection target, FIG. 12 is an explanatory diagram showing an example of the detection distance of the three-dimensional object for each section, and FIG. 13 is an example in which the detection distance of the three-dimensional object for each section is grouped FIG. 14 is an explanatory diagram showing the result of the detected object and side wall, FIG. 15 is an explanatory diagram showing an approximation of the contour of the own vehicle, FIG. 16 is an explanatory diagram showing the passing trajectory of the own vehicle, and FIG. FIG. 18 is an explanatory diagram of the determination area, and FIG. 18 shows the estimation result of the position of the three-dimensional object after the set time. 19 is an explanatory diagram showing tracking of out-of-field distance data, FIG. 20 is an explanatory diagram of measurement of a gap between a passing locus and distance data, and FIG. 21 is an explanatory diagram of contact determination between the host vehicle and a three-dimensional object. .
[0019]
In FIG. 1, reference numeral 1 denotes a vehicle such as an automobile, and the vehicle 1 is mounted with an out-of-vehicle monitoring device 2 that captures an image of an object within an installation range outside the vehicle and recognizes and monitors an object outside the vehicle from the captured image. Yes. The vehicle exterior monitoring device 2 includes a stereo optical system 10 for imaging an object outside the vehicle from different positions, an image processor 20 that processes an image captured by the stereo optical system 10 and calculates three-dimensional distance distribution information, The image processor 20 includes an image processing computer 30 that processes distance information from the image processor 20. The image processing computer 30 is connected to sensors for detecting the current running state of the vehicle, such as the vehicle speed sensor 4 and the steering angle sensor 5.
[0020]
The stereo optical system 10 is composed of a pair of left and right CCD cameras 10a and 10b using a solid-state image sensor such as a charge coupled device (CCD), for example. In the image processor 20, a pair of images captured by the CCD cameras 10a and 10b. The three-dimensional distance distribution over the entire image is calculated by a so-called stereo method, in which the correlation between the two images is obtained, and the distance is obtained from the parallax with respect to the same object by the principle of triangulation.
[0021]
The image processing computer 30 reads the distance distribution information from the image processor 20 to detect the road shape and the three-dimensional position of a plurality of three-dimensional objects (vehicles, obstacles, etc.) at high speed, The gap between the vehicle and the vehicle is calculated based on the change in position caused by the movement of the object and the traveling locus of the vehicle, and contact determination is performed. Numerical values and warnings are displayed on the display 9 installed in front of the driver. A video signal such as is sent to notify the driver.
[0022]
The image processor 20 and the image processing computer 30 have the hardware configuration shown in FIG. 2 in detail. The image processor 20 is a pair of stereo images captured by the CCD cameras 10a and 10b. A distance detection circuit that searches for a portion where the same object appears for each predetermined small area, calculates the distance to the object by calculating the amount of deviation of the corresponding position, and outputs it as three-dimensional distance distribution information 20a and a distance image memory 20b for storing distance distribution information output from the distance detection circuit 20a.
[0023]
The distance distribution information output from the distance detection circuit 20a is a pseudo image (distance image) shaped like an image, and is an image taken by the two left and right CCD cameras 11a and 11b, for example, as shown in FIG. When a long image (FIG. 8 schematically shows an image picked up by one camera) is processed by the distance detection circuit 20a, a distance image as shown in FIG. 9 is obtained.
[0024]
In the example of the distance image shown in FIG. 9, the image size is horizontal 600 pixels × vertical 200 pixels, and the distance data has a black dot portion. This is the horizontal direction among the pixels of the image of FIG. This is a portion where the change in brightness between pixels adjacent to each other is large. In the distance detection circuit 20a, this distance image is handled as an image composed of blocks of horizontal 150 × vertical 50 with a small area of 4 × 4 pixels, and the distance (number of pixel shifts) is calculated for each block. .
[0025]
On the other hand, the image processing computer 30 includes a microprocessor 30a mainly for detecting a road shape and the like, a microprocessor 30b mainly for detecting individual solid objects based on the detected road shape, and a detection. The microprocessor 30c mainly performs processing for determining a contact risk by identifying a preceding vehicle and an obstacle based on the position information of the three-dimensional object, calculating a gap with the own vehicle, and via the system bus 31. System configuration of multi-microprocessors connected in parallel.
[0026]
The system bus 31 stores an interface circuit 32 connected to the distance image memory 20b, a ROM 33 for storing a control program, a RAM 34 for storing various parameters in the middle of calculation processing, and parameters for processing results. I / O interface circuit for inputting signals from the output memory 35, the display controller (DISP.CONT.) 36 for controlling the display (DISP) 9, the vehicle speed sensor 4, the steering angle sensor 5, and the like. 37 is connected.
[0027]
As shown in FIG. 9, the image processing computer 30 treats the coordinate system on the distance image in units of pixels, with the lower left corner as the origin and the horizontal direction as the i coordinate axis and the vertical direction as the j coordinate axis. The point (i, j, dp) on the distance image where is set to dp is converted into a coordinate system in the real space, and processing such as road shape recognition and solid object position detection is performed.
[0028]
That is, the three-dimensional coordinate system of the real space is a coordinate system fixed to the own vehicle (vehicle 1), the X axis is the right side of the traveling direction of the vehicle 1, the Y axis is above the vehicle 1, and the Z axis is the vehicle 1 Assuming that the front and the origin are the road surface directly below the center of the two CCD cameras 10a and 10b, the XZ plane (Y = 0) coincides with the road surface when the road is flat. By the equations (1) to (3), the point (i, j, dp) on the distance image can be coordinate-converted to the point (x, y, z) on the real space.
x = CD / 2 + z · PW · (i-IV) (1)
y = CH + Z · PW · (j−JV) (2)
z = KS / dp (3)
However, CD: interval between CCD cameras 10a and 10b
PW: Viewing angle per pixel
CH: Mounting height of the CCD cameras 10a and 10b
IV, JV: Coordinates (pixels) on the image of the infinity point directly in front of the vehicle 1
KS: Distance coefficient (KS = CD / PW)
The equation for calculating the point (i, j, dp) on the image from the point (x, y, z) in the real space is obtained by modifying the above equations (1) to (3) as follows. .
[0029]
i = (x−CD / 2) / (z · PW) + IV (4)
j = (y−CH) / (z · PW) + JV (5)
dp = KS / z (6)
[0030]
Next, individual processing in the image processing computer 30 will be described. First, in the road detection processing by the microprocessor 30a, only the white line on the actual road is separated and extracted using the three-dimensional position information from the distance image stored in the distance image memory 20b and incorporated. The road shape is recognized by modifying / changing the parameters of the road model to match the actual road shape.
[0031]
In the above road model, the lane of the road up to the recognition target range is divided into a plurality of sections according to the set distance, and the left and right white lines are approximated by a three-dimensional linear equation for each section and connected in a polygonal line shape. The horizontal linear parameters a and b in the real space coordinate system and the vertical linear parameters c and d are obtained, and the horizontal linear equation shown in the following equation (7): Then, the vertical linear equation shown in the following equation (8) is obtained.
x = a · z + b (7)
y = c · z + d (8)
[0032]
In the three-dimensional object detection process by the microprocessor 30b, the distance image is divided into a grid pattern at predetermined intervals, and data above the road surface is determined for each of the sections based on the road shape detected by the microprocessor 30a. Are extracted as three-dimensional object data, and the distance to the three-dimensional object is calculated. Then, the sections of the detected three-dimensional object approaching each other are grouped into one group, and these groups are classified to detect structures along the road, such as the rear, side, and guardrail of the three-dimensional object. Then, the position and speed are calculated.
[0033]
Further, in the gap distance calculation process by the microprocessor 30c, the position of each section included in the detected three-dimensional object is moved according to the speed of the three-dimensional object, and the own vehicle detected by the vehicle speed sensor 4 and the steering angle sensor 5 is detected. The travel locus of the host vehicle is estimated from the travel state of the vehicle, and the position of the three-dimensional object after the movement is compared with the travel locus of the host vehicle to calculate the minimum gap distance.
[0034]
The generation of the distance image and the process of detecting the road shape from the distance image are described in detail in Japanese Patent Application Laid-Open Nos. 5-265547 and 6-266828 previously filed by the present applicant. ing.
[0035]
Hereinafter, the three-dimensional object detection process by the microprocessor 30b and the gap distance calculation process by the microprocessor will be described with reference to the flowcharts of FIGS.
[0036]
First, a three-dimensional object detection processing program by the microprocessor 30b will be described. In this program, first, in the distance data detection process in steps S101 to S114, the existence of a three-dimensional object and the calculation of the distance are calculated for each section obtained by dividing the distance image into a grid pattern at predetermined intervals. That is, when the road shape parameter is read in step S101, the distance image is divided into a grid pattern at predetermined intervals (for example, 8 to 20 pixel intervals) in step S102 as shown in FIG. The data of the three-dimensional object is extracted every time, and the data of the first section is read in order to calculate the detection distance.
[0037]
Next, when the process proceeds to step S104 and the first data in the section is set, the three-dimensional position (x, y, z) of the subject is obtained by the above-described equations (1) to (3) in step S105, and in step S106. Then, the height yr of the road surface at the distance z is calculated using the linear equations (7) and (8) of the road shape described above. Next, the process proceeds to step S107, and data above the road surface is extracted as three-dimensional object data based on the height H of the subject from the road surface calculated by the following equation (9).
H = y-yr (9)
[0038]
In this case, a subject whose height H is about 0.1 m or less is considered to be a white line, dirt, shadow, or the like on the road, so the data of this subject is rejected. An object above the height of the host vehicle 1 is also considered to be a pedestrian bridge, a sign, etc., so it is rejected and only the data estimated as a three-dimensional object on the road is selected.
[0039]
Thereafter, the process proceeds to step S108 to check whether it is the final data. If it is not the final data, the next data in the section is set in step S109 and the process returns to the above-mentioned step S105 to repeat the same processing above the road surface. Extract the data in Then, when the processing of the final data is completed within one section, the process proceeds from step S108 to step S110, and the number of data included in the section of the preset distance Z is counted for the extracted three-dimensional object data. Create a histogram with z as the horizontal axis.
[0040]
In subsequent step S111, a section where the frequency of the histogram is equal to or greater than the determination value and the maximum value is detected, and if there is a corresponding section, it is determined in step S112 that a three-dimensional object exists in the section, and the three-dimensional object is reached. Detect the distance.
[0041]
In the above histogram, there are some erroneously detected values in the distance data in the input distance image, and some data actually appears at positions where no object is present. However, when there is an object of a certain size, the frequency at that position shows a large value, whereas when there is no object, the frequency generated only by erroneous distance data becomes a small value. Therefore, if there is a section in which the frequency of the created histogram is greater than or equal to the preset judgment value and takes the maximum value, it is determined that an object exists in that section. If the maximum value of the frequency is less than the judgment value, the object exists. If the image data contains some noise, the object can be detected while minimizing the influence of the noise.
[0042]
Thereafter, the process proceeds from step S112 to step S113 to check whether or not the final section has been reached. When the final section has not been reached, the process proceeds from step S112 to step S114 and the next section of data is read. The solid object is detected and the distance is calculated in each section. The above process is repeated, and when the final section is reached, the process proceeds from step S113 to step S115.
[0043]
FIG. 12 is an example in which the image to be detected shown in FIG. 11 is detected by the above distance data detection process, and the distance for each section is shown on the XZ plane, and the detected distance includes some errors. Therefore, as indicated by black dots, the data is detected as data having some variation in the portion of the three-dimensional object facing the vehicle.
[0044]
These distance data are divided into groups that are close to each other by the distance group detection process in steps S115 to S121. In this process, the detection distance of the three-dimensional object in each section is examined, and if the difference in the detection distance to the three-dimensional object in the adjacent section is equal to or less than the determination value, it is regarded as the same three-dimensional object, while the determination value is exceeded Are grouped as if they were separate solid objects.
[0045]
For this reason, in step S115, first, the first section (for example, the left end) is examined. If a three-dimensional object is detected, the distance data is read and this section R1 is classified into group S1 and distance Z1. Next, proceeding to step S116, the right-side section R2 is examined, and if a three-dimensional object is not detected, the group S1 exists in and near the section R1, and the distance is determined to be Z1, If a three-dimensional object is detected in the section R2, and the detected distance is Z2, the difference between the distance Z1 in the section R1 and the distance Z2 in the right adjacent section R2 is calculated.
[0046]
Thereafter, the process proceeds to step S117 to check whether or not the difference in distance from the right adjacent section is equal to or less than the determination value. When the distance difference is equal to or less than the determination value and is approaching each other, in step S118, the difference is detected in section R2. The determined three-dimensional object is determined to belong to the previously detected group S1, and the same group is labeled. The distance is set as an average value of Z1 and Z2, and the process proceeds to step S120.
[0047]
On the other hand, when the difference in distance from the right adjacent section exceeds the determination value, the process proceeds from step S117 to step S119, and the three-dimensional object detected in section R2 is different from the previously detected group S1. Determination is made to label the new group (group S2, distance Z2), and the process proceeds to step S120.
[0048]
In step S120, it is checked whether or not the final section has been reached. If the final section has not been reached, the distance of the next section is read in step S121, the process returns to step S116, and the area on the right is checked. . When the final section is reached, the process proceeds from step S120 to step S122 and subsequent steps.
[0049]
By the distance group detection process described above, for example, distance data that are close to each other are collected from the distance data for each section shown in FIG. 12, and are grouped as shown in FIG. At this time, the number of the group to which the category belongs is assigned and stored in each category. In the example of FIG. 13, groups 1, 2 and 3 are assigned to the group to which the section having the implantation distance data on the left in the traveling direction belongs, and the group having the distance data of the rear part of the preceding vehicle belongs to the group to which the group belongs. Group 4 is numbered. Further, the group 6 is assigned to the group to which the section having the distance data of the right bicycle in the traveling direction belongs, and the group 7 is assigned to the group to which the section having the distance data of the right wall belongs.
[0050]
In this case, even if they are different three-dimensional objects, sections that are close to each other may be processed as the same group. For this reason, in the group division processing in the next step S122 to step S132, the arrangement direction of the distance data on the XZ plane is examined, and the arrangement direction is substantially parallel to the Z axis and the substantially parallel part to the X axis. Divide the group with.
[0051]
In this group division process, in step S122, the first group of data is read, and in step S123, the alignment direction of each section in the group is calculated. In step S124, the labels of “object” and “side wall” are assigned to each section. Add. Specifically, assuming that the position of the leftmost section K1 in the group is Z1 and X1 and the positions of the rightmost sections K are Zp and Xp, two points of points X1 and Z1 and points Xp and Zp are obtained. An inclination A1 of the connecting straight line with respect to the Z axis is calculated, and the inclination A1 of the straight line is compared with a set value (for example, about 45 °). If the slope A1 of the straight line is equal to or less than the set value and the data sequence is substantially in the Z-axis direction, the section K1 is labeled “side wall”, the slope A1 of the straight line exceeds the set value, and the data sequence is In the case of a substantially X-axis direction, it is labeled “object”.
[0052]
The interval N of the division at the time of labeling is about N = 2 to 4 divisions. This is because in N = 1, that is, in the right adjacent section, the arrangement direction greatly varies due to variations in the detection distance, and it is difficult to determine the division. By using the arrangement direction with a slightly separated section, , To stabilize the direction. Then, the labeling of the “side wall” or “object” is performed in order from the left end in the group from the right end to the left N sections, and each section is labeled.
[0053]
As described above, when labeling of each section is completed, the process proceeds from step S124 to step S125, the label of the leftmost section is read, and further, the label of the next right section is read in step S126. Next, the process proceeds to step S127, and it is checked whether the leftmost label is different from the label on the right. As a result, in step S127, when the labels are the same, the process jumps to step S129, and when the labels are different, the section labeled “side wall” and the section labeled “object” are divided in step S128. As another group, the process proceeds to step S129. The position of the segment to be divided is on the right side by N / 2 segments where the label changes from “side wall” ← → “object”.
[0054]
In this case, in order to deal with a situation where the label partially changes due to variations in distance data, etc., the division is performed only when the same label is continuously arranged more than the judgment value (for example, three or more categories). If it is less than 1, no division is performed. For example, in the example of FIG. 13, as in the case of group 6 (bicycle) or the like, a group in which the section labeled “object” and the section labeled “side wall” are mixed because the shape is complicated. Then, since the same type of labels do not continue in three or more sections, group division does not occur.
[0055]
In step S129, it is checked whether or not it is the final division. If it is not the final division, the label of the next division is read in step S130, the flow returns to step S126, and the same processing is repeated. When the final section is reached, the process proceeds from step S129 to step S131 to check whether the final group has been reached. As a result, when the final group has not been reached, the data of the next group is read in step S132, and the group is similarly divided into the next group. This process is repeated, and when the final group is reached, the group division process is completed and the process proceeds from step S131 to step S133.
[0056]
The next step S133 to step S138 are processing for calculating the parameters of each group by classifying each divided group as a side wall or an object. When the first group data is read in step S133, In S135, an approximate straight line is obtained from the position (Xi, Zi) of each section in the group by the Hough transform or the least square method, and the inclination of the entire group is calculated.
[0057]
Next, proceeding to step S135, the group having the inclination in the X-axis direction is classified as an object and the group having the inclination in the Z-axis direction is classified as a side wall from the inclination of the entire group, and the parameters of each group are calculated in step S136. To do. This parameter is a parameter such as an average distance calculated from the distance data in the group and the X coordinate of the left end and the right end in a group classified as an object. In a group classified as a side wall, the alignment direction (Z Parameters such as the inclination with respect to the axis and the positions of the front and rear ends (Z, X coordinates).
[0058]
The group classification may be performed based on the label of “side wall” or “object” of each section attached in the group division process described above.
[0059]
Then, the process proceeds from step S136 to step S137 to check whether or not the final group has been reached.When the final group is not reached, the next group data is read in step S138 and the process returns to step S134. It progresses to the process after step S139.
[0060]
Here, one surface of a three-dimensional object, such as a continuous guard rail, may be strongly affected by the variation in distance data for each section, and is divided into a plurality of groups by the previous distance group detection process or group division process. Therefore, there is a possibility that one surface of the three-dimensional object is erroneously divided and detected.
[0061]
As a countermeasure against this, in the group joining process in the following steps S139 to S147, the mutual positional relationship of each group is examined based on the group parameter calculated in the previous group parameter calculation process, and the position of the end point approaches in the same type of group. If the arrangement directions are substantially equal, it is determined that they are the same surface of the same object, and these groups are integrated into one group. Then, various parameters as an integrated group are recalculated in the same manner as the group parameter calculation process.
[0062]
Therefore, in step S139, the parameters of the first group are read, and in step S140, the parameters of the next group are read.In step S141, the difference in the distance between the end points of each group and the difference in the inclination of each group are obtained. calculate. In step S142, it is checked whether the difference in the distance between the end points of each group and the difference in the inclination of each group are within the respective determination values. If both are within the determination values, the process proceeds to step S143 and the same object The groups are joined as a group, the group parameters are calculated again, and the process proceeds to step S146.
[0063]
On the other hand, if it is determined in step S142 that the distance difference between the end points of each group or the difference in slope of each group is not within the determination value, the process proceeds from step S142 to step S144 to check whether the group is the final group. If not, the parameter of the next group is read in step S145, and the process returns to step S141. If it is the final group, the process proceeds to step S146.
[0064]
In step S146, it is checked whether or not it is the final group. If it is not the final group, the parameters of the next group are read in step S147 and the process returns to step S140. If it is the final group, the process proceeds to step S148 and subsequent steps.
[0065]
After the above group combining process, the same three-dimensional object is detected by the processes of steps S148 to S157 below for those separated into different groups at the rear and side surfaces of the same three-dimensional object, One solid object is recognized as a combination of an “object” and a “side wall” (the rear part is “object” and the side surface is “side wall”).
[0066]
In this process, first, in step S148, parameters of the first group classified as an object are read, and in step S149, parameters of the first group classified as a side wall are read. Next, the process proceeds to step S150, and the difference between the position of the end point of the group classified as an object and the position of the end point of the group classified as a side wall is calculated. In this case, when the “object” is on the right side of the front of the automobile (corresponding to the Z axis), the position of each end point is the position of the left end of the “object” and the position of the end point on the near side of the “side wall”. The difference is calculated, and if the “object” is on the left side of the front of the host vehicle, the difference between the right end position of the “object” and the end point position on the near side of the “side wall” is calculated.
[0067]
Then, in step S151, it is checked whether or not the difference between the positions of the end points of each group is within a determination value (for example, about 1 m), and whether or not they are close to each other. If it is not the final side wall, the parameter of the next group classified as the side wall in step S153 is read and the process returns to step S150. If it is the final side wall, the process proceeds to step S156.
[0068]
On the other hand, when the difference between the positions of the end points of each group is within the determination value in step S151, the process proceeds from step S151 to step S154 and is determined to be the same three-dimensional object. In other words, when the rear and side surfaces of a solid object are visible at the same time, the corner created by the two surfaces is convex forward, so the left end position of the “object” and the end point of the front side of the “side wall” Or the position of the right end of the “object” and the position of the end point on the near side of the “side wall” are essentially the same. Therefore, when the difference between the positions of the two groups is within the determination value, it can be determined that the two groups are pairs detected by dividing one solid object.
[0069]
Thereafter, the process proceeds from step S154 to step S155, and a point that intersects the pair of “object” and “side wall” determined to be the same three-dimensional object by extending each approximate straight line is determined as the corner point of the three-dimensional object. When calculated as a position, the position of each end point is changed to the position of this corner point. Then, in step S156, it is checked whether or not it is the final group of “object”. If it is not the final group of “object”, the parameter of the next group classified as “object” is read in step S157, and the process proceeds to step S149. Return and repeat the same process. On the other hand, if it is the last group of “objects” in step S156, the process proceeds to the next step S158 and subsequent steps.
[0070]
FIG. 14 shows the detection results of “object” and “side wall” in the XZ plane with respect to the detection target example shown in FIG. 11. In this example, the above-described group combination and “object” are shown. There is no group that is determined to be a combination of “side wall”, “object” is indicated by a bold solid line, and “side wall” is indicated by a thick dashed line.
[0071]
Next, in step S158 and subsequent steps, a speed calculation process for calculating the moving speed based on a change in the position of the “object” or “side wall” detected every processing cycle at a predetermined time interval (for example, 0.1 sec). First, in step S158, the parameters of the first group are read, and in step S159, it is checked whether or not they are a pair of the same three-dimensional object.
[0072]
When the same solid object is not paired, the process proceeds from step S159 to step S160. The center of the object is the center of the left and right ends of the “object”, and the center of the front and rear ends of the “side wall”. When the time change amount of the position is calculated, in step S161, the speed in the front-rear direction, that is, the speed Vz in the Z direction is calculated from the time change amount of the Z coordinate of the center point, and the speed in the left-right direction, that is, the speed in the X direction is Calculation is made from the amount of time change of the X coordinate of the center point, and the process jumps to step S164.
[0073]
In step S164, it is checked whether or not it is the final group. If it is not the final group, the parameters of the next group are read in step S165, and the process returns to step S159 to check whether or not it is a pair of the same three-dimensional object. As a result, when the pair is not a pair of the same three-dimensional object, the speed in the front / rear / left / right directions is calculated from the time change at the center point through the above-described steps S160 and S161, and the process returns to step S164 to check whether it is the final group.
[0074]
On the other hand, when the same three-dimensional object is a pair in step S159, the process proceeds from step S159 to step S162, and the corresponding “object” or “side wall” parameter is read. While calculating the velocity (velocity Vz in the Z-axis direction), the velocity in the left-right direction (velocity Vx in the X-axis direction) is calculated from the “side wall”, and these velocity Vz, Vx are expressed in the Z-axis direction in the same three-dimensional object, The speed is in the X-axis direction. Then, in step S164, it is checked whether or not it is the final group. If it is not the final group, the above processing is repeated. If it is the final group, the process proceeds from step S164 to step S166, and the parameters of each group are written in the memory. The entire detection process program is terminated.
[0075]
That is, in the “object”, the position in the Z-axis direction is an average value of the distances of the plurality of sections, and the velocity Vz is stable, but the position in the X-direction is affected by the variation of the X coordinate at the left and right ends. The speed Vx tends to vary greatly. On the other hand, in the “side wall”, the position in the X direction is an average value of the X coordinates of a plurality of sections, and the velocity Vx is stable, but the position in the Z direction is affected by variations in the Z coordinates at the front and rear ends. As a result, the speed Vz tends to vary greatly. Therefore, the speed of the same three-dimensional object is set by using only both stable speeds.
[0076]
In this case, it is difficult to obtain the correspondence relationship between the distance data detected in the previous process and the distance data detected in the current process and the distance data detecting the same part of the same three-dimensional object. It is difficult to obtain the speed of an object from individual distance data. However, in the present invention, individual distance data is grouped for each three-dimensional object, approximated linearly for each group, and the center point is obtained, so the group detected at the previous processing and the current processing are It is easy to obtain the correspondence relationship with the detected group, and the speed of the object can be calculated from the movement amount of the center point.
[0077]
The object and side wall data detected by the above processing is passed to the clearance distance calculation processing program by the microprocessor 30c, and the contact risk based on the position of the three-dimensional object after the prediction time t and the traveling area of the host vehicle. Is determined, and the gap distance between the host vehicle and the three-dimensional object is calculated.
[0078]
In this clearance distance calculation processing program, the trajectory of the host vehicle is estimated in step S201. The passing trajectory of the host vehicle is estimated using a built-in model in which the contour of the host vehicle is linearly approximated by a polygon composed of n vertices. As shown in FIG. 15 (b), the contour is the coordinates Pn = [Pxn, Pzn] of the vertex n with the rear axle center as the origin. ^T And the coordinates Pn ′ = [P′xn, P′zn] of the vertex n after the prediction time t (for example, 1 or 2 seconds) ^T Is obtained to estimate the passing trajectory of the host vehicle.
[0079]
For this reason, first, the idle running distance L and the turning radius R based on the rear axle center at the predicted time t are obtained from the vehicle speed obtained from the vehicle speed sensor 4 and the steering angle obtained from the steering angle sensor 5. This idle travel distance L is composed of a distance proportional to the speed of the host vehicle and a minimum travel distance (for example, about ˜1 m), and the center of the turning radius R overlaps the rear axle extension line. And the coordinates of the turning center are O = [Ox, Oz] ^T Assuming that the turning angle is θ, the turning angle θ is given by the following equation (10), and the coordinates of the vertex n after the prediction time t are given by the following equation (11).
θ = L / R (10)

[0080]
Moreover, since the turning radius rn of each vertex can be obtained by the following equation (12), using this, the vertex number no having the largest turning radius (passing through the outermost side) and the smallest (passing through the innermost side) are used. The vertex number ni is obtained. The current coordinate Pno of the vertex number no and the coordinate P′no after the prediction time t are connected by an arc having the maximum turning radius rno obtained by applying the equation (12), and the current coordinate of the vertex number ni. By connecting Pni and the coordinates P′ni after the predicted time t by an arc having the minimum turning radius rni obtained by applying the equation (12), the remaining vertices are connected by a straight line to each adjacent vertex. As shown in FIG. 16, the passing trajectory of the host vehicle can be obtained.
rn = ((Pxn-Ox) ² + (Pzn-Oz) ² ) ^1/2 … (12)
[0081]
In this case, when the steering angle is 0, that is, when the turning radius is infinite, the calculation cannot be performed. Therefore, the coordinates of the vertex n after the prediction time t are obtained by the following equation (13) and Are connected by a straight line to obtain an estimated trajectory.

[0082]
Then, after estimating the passing trajectory of the host vehicle, the process proceeds from step S201 to step S202, and the center of the tip of the contour model of the host vehicle (see FIG. 15) Cf = [Cxf, Czf for contact determination and gap measurement described later. ] ^T And rear center Cr = [Cxr, Czr] ^T As in the case of each vertex, the positions C′f and C′r after the prediction time t are calculated, and as shown in FIG. 17, the travel area based on the passage trajectory is divided into a right determination area and a left determination area. To do.
[0083]
In the subsequent step S203, the X-direction movement amount dx and the Z-direction movement amount dz of the three-dimensional object at the prediction time t are calculated from the velocities Vx and Vz of each group obtained in the previous three-dimensional object detection process by the following (14), ( The distance data is obtained by the equation (15), and each distance data is moved by the X-direction movement amount dx and the Z-direction movement amount dz to estimate the position of the three-dimensional object after the prediction time t.
dx = Vx · t (14)
dz = Vz · t (15)
[0084]
In the example in which the detection results of the positions of the three-dimensional objects shown in FIG. 13 are grouped, the velocity is substantially 0 in both the X direction and the Z direction in the groups 1, 2, and 3 in which the implantation is detected and the group 7 in which the right wall is detected. Therefore, the position after the prediction time t does not change from the detection position. Further, in the group 4 in which the preceding vehicle is detected, for example, assuming that the speed Vx = 0 and Vz = 4.0 m / sec are detected, and t = 1.5 sec, the movement amount of the group 4 is dx = 0. 0.0 m and dz = 6.0 m. For example, if the group 6 that detected the bicycle is detected as Vx = −0.1 m / sec and Vz = −1.0 m / sec, the movement amount is dx = −0.15 m, dz = −1. .5m. FIG. 18 shows an estimated position after the prediction time t when each distance data is moved by the X-direction movement amount dx and the Z-direction movement amount dz.
[0085]
Thereafter, the process proceeds to step S204, and distance data is tracked for a three-dimensional object outside the field of view of the stereo camera. The group of distance data detected as having a velocity of approximately 0 in both the X direction and the Z direction is a three-dimensional object stationary on the road, and the distance data is held for each processing cycle.
[0086]
A travel locus between the k-th process cycle and the (k−1) -th process cycle can be obtained as a travel locus L ′ and a turning radius R ′ by the vehicle speed sensor 4 and the steering angle sensor 5. At this time, since the travel distance L ′ is based on the position in the k-th processing cycle, L ′ <0, and the coordinates of the turning center are O ′ = [O′x, O′z]. ^T Assuming that the turning angle is α, the turning angle α is given by the following equation (16).
α = L '/ R' (16)
[0087]
Therefore, the distance data in the (k-1) th processing cycle is represented by D = [Dx, Dz]. ^T Then, the distance data D ′ in the k-th processing cycle can be obtained by the following equation (17), and by updating this every processing cycle, as shown in FIG. It is possible to track distance data that has deviated outside the field of view.

[0088]
After that, in order to determine contact between the vehicle passing trajectory and the three-dimensional object, including the tracking data of the three-dimensional object that has deviated from the field of view, in step S205, the distance data in which the position after the prediction time t is estimated In step S206, it is checked whether or not the estimated distance data is within the right determination area.
[0089]
Then, if the estimated distance data is not within the right determination area, the process proceeds from step S206 to step S208. If the estimated distance data is within the right determination area, it is determined that there is a possibility of contact with the own vehicle, and from step S206. Proceeding to step S207, the right contact flag indicating right contact is turned ON, and the process proceeds to step S208.
[0090]
In step S208, it is checked whether the estimated distance data is in the left determination area. If the estimated distance data is not in the left determination area, the process proceeds to step S210. If the estimated distance data is in the left determination area, the left distance is determined in step S211. The left contact flag indicating contact is turned ON, and the process proceeds to step S210.
[0091]
In step S210, it is checked whether the right or left contact flag is turned on. If the contact flag on either side is turned on, an alarm is generated in step S211 and a warning is displayed on the display 9 or a lamp is displayed. Warning by buzzer etc. and inform the driver.
[0092]
On the other hand, when the left and right contact flags are not turned on in step S210, the process proceeds from step S210 to step S212, and the estimated distance data outside the contact determination area is determined based on the vehicle center except for the data in front of the front of the passage locus. Divide the line left and right into groups. Next, in step S213, the gap distance between the estimated distance data outside the contact determination area and the passing trajectory of the host vehicle is calculated. As shown in FIG. 20, this gap distance draws a normal line with respect to the passing trajectory on the left side surface of the host vehicle in the case of the left group and the passing trajectory on the right side surface of the host vehicle in the case of the right group. The value is stored as the “gap distance”.
[0093]
Thereafter, the process proceeds to step S214 to check whether or not the final estimated distance data is reached. When the final estimated distance data is not reached, the next estimated distance data is read in step S215, and the process returns to step S206. repeat. When the contact determination is completed for all estimated distance data, the process proceeds from step S214 to step S216, and the minimum value of the gap distance is calculated and held for each of the left and right groups, and the numerical value displayed or calculated gap distance in step S217 A display signal for calling attention according to the minimum value is output to the display 9 to notify the driver.
[0094]
FIG. 21 shows the travel region of the host vehicle for the estimated position of each distance data after the predicted time t shown in FIG. 18, and in this example, the current detected position of the preceding vehicle is the travel of the host vehicle. Although it is in the area, the contact determination is performed for the estimated position after the set time in consideration of the speed of the three-dimensional object and the traveling locus of the host vehicle, so that it can be correctly determined that the vehicle does not contact the preceding vehicle. .
[0095]
In addition, even for a complex object such as a bicycle in FIG. 21, since the individual distance data in the group is moved according to the speed of the group, the unevenness of the detected object is maintained as it is, and contact determination is performed. Can be performed precisely. In the example of FIG. 21, it is correctly determined that there is a risk of contact with the bicycle, and an unexpected accident can be avoided in advance. In FIG. 21, the same applies to a three-dimensional object that is stationary, such as implantation.
[0096]
In addition, in this case, in order to estimate the travel locus drawn by the contour of the host vehicle according to the speed and the steering angle of the host vehicle, in addition to considering the unevenness of the object, the unevenness of the host vehicle is also considered, and the contact determination In addition to improving the accuracy of gap measurement, the driver's gaze area can be brought closer to the contact determination area, and an alarm with less discomfort can be issued.
[0097]
Furthermore, since distance data is tracked even for an object that has deviated outside the detection range, contact determination and gap measurement can be performed on a solid object immediately next to the host vehicle. It can exert a great effect in preventing contact accidents due to the influence of the inner ring difference.
[0098]
FIGS. 22 and 23 relate to the second embodiment of the present invention, FIG. 22 is a partial flowchart of the three-dimensional object detection process, and FIG. 23 is an explanatory diagram showing the correspondence between groups.
[0099]
In the first embodiment described above, in order to detect the moving speed of the three-dimensional object, the detected group is approximated in a straight line as an “object” or “side wall”, and the center point is obtained. On the other hand, in this embodiment, the individual distance data in the group detected in the previous process is compared with the individual distance data in the group detected in the current process, the movement amount as a group is obtained, and the speed is calculated. Is.
[0100]
For this reason, in this embodiment, a part of the three-dimensional object detection process by the microprocessor 30b of the image processing computer 30 is changed. The three-dimensional object detection process of the present embodiment is the distance data detection process of steps S101 to S114, the distance group detection process of steps S115 to S121, and the group division process of steps S122 to S132 in the program of FIGS. The process is the same until the group division process is changed.
[0101]
In other words, in this embodiment, when the group division process of steps S122 to S132 is completed, the process proceeds from step S132 to step S170 of FIG. 22, and in steps S170 to S174, the current process is selected from the plurality of groups detected in the previous process. After searching for a group that detects the same three-dimensional object with the group detected in step 1 and performing a process for obtaining the corresponding relationship of the group, in step S175 and subsequent steps, there is a corresponding relationship with the group detected in the current process. The speed of the group is obtained from the amount of movement with the previously detected group.
[0102]
Therefore, first, when the data of the first group detected this time is read in step S170, the center point (Xc, Zc) of the distance data in the group is obtained in step S171. The barycentric point (Xc, Zc) is expressed by the following equations (18) and (19), where xi and zi are the positions of the i-th distance data in the group and n is the number of distance data belonging to the group. Can be sought. Here, Σ is the sum of i = 1 to n.
Xc = (Σxi) / n (18)
Zc = (Σzi) / n (19)
[0103]
Next, proceeding to step S172, the barycentric point of the group detected in the previous process is read, and the distance (the amount of movement of the barycentric point) between the barycentric point of the previously detected group and the barycentric point of the currently detected group is calculated. After the calculation, in step S173, it is checked whether or not the distance between the centroid points of each group is equal to or less than the determination value. When the distance between the center of gravity of the group detected in the previous process and the center of gravity of the group detected this time exceeds the determination value, it is determined that these groups do not detect the same three-dimensional object. Then, the process proceeds to step S174, the barycentric point of the next group detected last time is read, and the process returns to step S172.
[0104]
On the other hand, when the distance between the centroids of each group is equal to or smaller than the determination value in step S173, it is determined that these groups are in a correspondence relationship in which the same solid object is detected, and the process proceeds from step S173 to step S175. Each distance data in the detection group is moved to calculate a correction amount that minimizes the shift amount between the groups.
[0105]
For example, as shown in FIG. 23, if the previously detected group F and the currently detected group G are in a correspondence relationship, first, each distance data of the previously detected group F is obtained from the center point PG of the currently detected group G. And the center-of-gravity point PF of the previously detected group F move.
[0106]
Next, for each distance data iF1 of the group F1 after the movement, distance data iG that is closest in the group G detected this time is obtained, and a distance diF1 between the distance data iF1 and iG is obtained. This is performed for all the distance data in the group F1 after movement, and the square sum SDK1 (deviation amount) of the distance diF1 is obtained by the following equation (20).
SDK1 = ΣdiF1 ² … (20)
[0107]
Further, the entire position of the group F1 is moved in the Z direction and the X direction by minute amounts Δz and Δx, and the deviation amount SDK2 between the group F2 after this movement and the group G detected this time is set in the same manner as described above. calculate. Then, this process is repeated while changing Δz and Δx, and the values of the minute amounts Δz and Δx from which the minimum value is obtained among the deviation amounts SDK1, SDK2, SDK3,. The amount of correction becomes.
[0108]
As described above, when the correction amount that minimizes the amount of deviation between the groups is obtained, the process proceeds to step S176, and the value obtained by adding the correction amount to the movement amount of the center of gravity between the groups in the correspondence relationship is obtained. In step S177, the movement speed is divided by the time interval for each processing cycle to calculate the group speeds Vz and Vx. Next, in step S178, it is checked whether or not it is the final group. If it is not the final group, the data of the next group is read in step S179, the process returns to step S171, and the above processing is repeated. For simplicity, the moving speed of the center of gravity calculated in step S172 may be divided by the time interval for each processing cycle to obtain the group speed.
[0109]
Then, when the final group processing is completed in step S178, the process proceeds from step S178 to step S180, and parameters such as the position and speed of each group are written in the memory, and the entire three-dimensional object detection processing is performed. Exit the program. The subsequent gap distance calculation process using the data of the three-dimensional part detection process is the same as that in the first embodiment.
[0110]
In this embodiment, the moving speed of the three-dimensional object can be detected with high accuracy compared to the first embodiment described above, and the accuracy of contact determination and gap distance can be further improved.
[0111]
FIGS. 24 to 27 relate to a third embodiment of the present invention, FIG. 24 is an overall configuration diagram of the outside monitoring apparatus, FIG. 25 is a circuit block diagram of the outside monitoring apparatus, and FIG. 26 is an explanatory view showing an example of an image. These are explanatory drawings showing an example of a two-dimensional distribution of a three-dimensional object.
[0112]
As shown in FIG. 24, the vehicle exterior monitoring device 50 according to the present embodiment processes an image from a stereo optical system 60 including two upper and lower cameras to recognize a two-dimensional position distribution of a three-dimensional object. 70, and this passage pattern recognition device 70 is connected to the same image processing computer 30 as in the first embodiment. The position information of the two-dimensional distribution of the three-dimensional object obtained by the passage pattern recognition device 70 is described above. Processing is performed in the same manner as in the first embodiment.
[0113]
As the passage pattern recognition device 70, a known device can be applied. For example, the Society of Instrument and Control Engineers Vol. 21, no. 2 (February 1985) The passage pattern recognition device described in “Two-dimensional distribution recognition method of obstacle” can be applied. The passage pattern recognition device 70 detects the distance distribution to the subject by processing the images of the two cameras as in the first embodiment, but the data processing method inside the device is different. Only the information of the three-dimensional object is output, and the road shape cannot be detected by the white line as in the first embodiment.
[0114]
The stereo optical system 60 is composed of two CCD cameras 60a and 60b that are attached to the front portion of the vehicle 40 with a certain vertical distance, and the passage pattern recognition device 70 is as shown in FIG. The upper and lower two image signals from the stereo optical system 60 are input, the distance detection circuit 60a for calculating the two-dimensional distribution of the distance and position of the three-dimensional object, and the two-dimensional distribution information from the distance detection circuit 60a are stored. It is composed of a two-dimensional distribution memory 60b and the like.
[0115]
In this embodiment, when an image captured by the stereo optical system 60, for example, an image as shown in FIG. 26 is processed by the passage pattern recognition device 70, a two-dimensional distribution pattern of a three-dimensional object as shown in FIG. 27 is output. This corresponds to the distance data for each section in the first form described above. By performing the same processing as in the first form on the two-dimensional distribution pattern of this three-dimensional object, the three-dimensional object detection, Contact determination and gap distance measurement can be performed.
[0116]
In this embodiment, since the presence / absence and position of the side wall are converted into a simple data form such as a linear parameter and front / rear end coordinates, data handling and processing on the use side is facilitated.
[0117]
FIGS. 28 to 32 relate to a fourth embodiment of the present invention, FIG. 28 is an overall configuration diagram of the outside monitoring apparatus, FIG. 29 is a circuit block diagram of the outside monitoring apparatus, and FIG. 30 is a side view of a laser beam scanning method. FIG. 31 is an explanatory view showing a laser beam scanning method from above, and FIG. 32 is an explanatory view showing an example of a two-dimensional distribution of a three-dimensional object measured by a laser radar distance measuring device.
[0118]
In each of the above-described embodiments, an image outside the vehicle is detected by processing an image captured by the camera. On the other hand, this embodiment detects an object outside the vehicle by scanning a laser beam. That is, as shown in FIG. 28, the vehicle outside monitoring device 90 mounted on the vehicle 80 of this embodiment employs a laser radar distance measuring device 100 using a laser beam, and the image processing computer 30 is used in the laser radar distance measuring device 100. Are connected.
[0119]
The laser radar distance measuring device 100 projects a laser beam, receives light reflected by the laser beam and hits an object, and measures the distance from the required time to the object. A known laser radar distance measuring device can be applied to the device 90.
[0120]
In the outside monitoring apparatus 90 of this embodiment, a laser projection unit 101 having a laser beam projection / light reception function and a horizontal scanning function is attached to the front portion of the vehicle. As shown in FIG. 29, laser radar measurement is performed. In the distance device 100, a distance detection circuit 100a that calculates the distance to the object from the time required for projecting and receiving the laser beam and that calculates the two-dimensional position of the object from the scanning direction of the laser beam is detected. It consists of a two-dimensional distribution memory 100b for writing the two-dimensional position of the object.
[0121]
As shown in FIG. 30, a laser beam is projected horizontally from the laser projection unit 101, and only a three-dimensional object at a position higher than the road surface is detected. In addition, as shown in FIG. 31, the laser beam is scanned in the left-right direction, and the operation of detecting the distance by projecting and receiving the laser beam at a predetermined interval in a predetermined scanning range is repeated, so that two-dimensional objects are obtained. A dimensional distribution is measured.
[0122]
For example, when the laser radar distance measuring device 100 measures the situation where there is a guard rail on the left front side and another vehicle on the right front, information on a two-dimensional distribution of a three-dimensional object as shown in FIG. 32 is obtained. This is the same as the distance data of the three-dimensional object for each section in the first embodiment.
[0123]
Therefore, an object or a wall surface can be detected by performing the same processing as the first embodiment on the two-dimensional distribution of the three-dimensional object that is the output of the laser radar distance measuring device 100. In this embodiment, data of a three-dimensional object can be obtained in a form that can be easily processed, and the amount of calculation processing can be reduced.
[0124]
【The invention's effect】
As explained above, according to the present invention, , Inspection It is possible to obtain an excellent effect such that the possibility of contact with the host vehicle can be accurately and precisely determined by taking in information on the position change of the obstacle and the detailed shape.
[Brief description of the drawings]
FIG. 1 relates to a first embodiment of the present invention and is an overall configuration diagram of a vehicle outside monitoring device.
FIG. 2 is a circuit block diagram of the vehicle exterior monitoring device.
FIG. 3 is a flowchart of a three-dimensional object detection process (No. 1).
FIG. 4 is a flowchart of a three-dimensional object detection process (No. 2).
FIG. 5 is a flowchart of a three-dimensional object detection process (No. 3).
FIG. 6 is a flowchart of a three-dimensional object detection process (No. 4).
FIG. 7 is a flowchart of a gap distance calculation process as described above.
FIG. 8 is an explanatory diagram showing an example of an image captured by the in-vehicle camera.
FIG. 9 is an explanatory diagram showing an example of a distance image
FIG. 10 is an explanatory diagram showing divisions of distance images as above.
FIG. 11 is an explanatory diagram showing an example of the situation of the detection target
FIG. 12 is an explanatory diagram showing an example of the detection distance of a three-dimensional object for each section.
FIG. 13 is an explanatory diagram showing an example in which detection distances of three-dimensional objects for each section are grouped as above.
FIG. 14 is an explanatory diagram showing the result of the detected object and the side wall, same as above.
FIG. 15 is an explanatory diagram showing approximation of the contour of the own vehicle
FIG. 16 is an explanatory diagram showing the passing trajectory of the own vehicle
FIG. 17 is an explanatory diagram of the left / right determination area.
FIG. 18 is an explanatory diagram showing the estimation result of the position of the three-dimensional object after the set time;
FIG. 19 is an explanatory diagram showing tracking of distance-of-view distance data as above.
FIG. 20 is an explanatory diagram of the measurement of the gap between the passage locus and the distance data.
FIG. 21 is a diagram for explaining the contact determination between the host vehicle and the three-dimensional object.
FIG. 22 is a partial flowchart of a three-dimensional object detection process according to the second embodiment of the present invention.
FIG. 23 is an explanatory diagram showing the correspondence of groups as in the above.
FIG. 24 relates to a third embodiment of the present invention, and is an overall configuration diagram of a vehicle exterior monitoring device;
FIG. 25 is a circuit block diagram of the vehicle outside monitoring device.
FIG. 26 is an explanatory diagram showing an example of an image as above;
FIG. 27 is an explanatory diagram showing an example of a two-dimensional distribution of a three-dimensional object.
FIG. 28 relates to a fourth embodiment of the present invention and is an overall configuration diagram of a vehicle outside monitoring device.
FIG. 29 is a circuit block diagram of the vehicle exterior monitoring device as above.
FIG. 30 is an explanatory diagram showing the laser beam scanning method from the side, same as above.
FIG. 31 is an explanatory view showing the laser beam scanning method from above,
FIG. 32 is an explanatory diagram showing an example of a two-dimensional distribution of a three-dimensional object measured by the laser radar range finder;
[Explanation of symbols]
1 ... Outside monitoring device
10a, 10b ... CCD camera
20 Image processor
30. Image processing computer

Claims (7)

In the outside monitoring device that recognizes the situation outside the vehicle by dividing the pseudo image of the distance distribution obtained by measuring the distance of the object outside the vehicle into a plurality of sections and detecting the presence or absence of a three-dimensional object and distance data for each section ,
Means for calculating the moving speed of a three-dimensional object for each group by grouping the adjacent sections in which the distance data is close to each other into one group;
Means for moving the distance data of all sections belonging to the group according to the moving speed of the group to which the section belongs, and estimating the position of the three-dimensional object after a set time;
Means for estimating a travel locus drawn by the contour of the vehicle after the set time;
Means for comparing the estimated position of the three-dimensional object with the traveling locus of the host vehicle and determining whether the host vehicle is in contact with the three-dimensional object;
A vehicle exterior monitoring device comprising: means for calculating a minimum clearance between the host vehicle and the three-dimensional object based on the estimated position of the three-dimensional object and the travel locus of the host vehicle.   Compare the distance data in the group detected in the previous process with the distance data in the group detected in the current process to check the correspondence between groups, and the amount of time movement between the corresponding groups The vehicle outside monitoring device according to claim 1, wherein the moving speed is calculated from the vehicle.   The distance data of the three-dimensional object that is stationary on the road is held for each processing cycle, and when the three-dimensional object deviates from the detection range, the three-dimensional object 2. A vehicle exterior monitoring device according to claim 1, further comprising means for estimating a position.   2. The pseudo image according to claim 1, wherein a correlation between a pair of images taken by a pair of cameras is obtained and a three-dimensional distance distribution obtained by a triangulation principle from parallax with respect to the same object. The vehicle exterior monitoring device described.   2. The out-of-vehicle monitoring apparatus according to claim 1, wherein the pseudo image shows a two-dimensional distance distribution obtained by processing a pair of images captured by a set of cameras.   2. The vehicle exterior monitoring apparatus according to claim 1, wherein the pseudo image shows a two-dimensional distance distribution obtained by projecting / receiving light of a laser beam or transmitting / receiving a radio wave. In the outside monitoring device that measures the distance of an object outside the vehicle, detects the presence or absence of a three-dimensional object and distance data for the three-dimensional object, and recognizes the situation outside the vehicle,
Means for gathering the distance data adjacent to and close to each other into one group and calculating the moving speed of the three-dimensional object for each group;
Means for moving the distance data of all sections belonging to the group according to the moving speed of the group to which the section belongs, and estimating the position of the three-dimensional object after a set time;
Means for estimating a travel locus drawn by the contour of the vehicle after the set time;
A vehicle exterior monitoring apparatus comprising: means for comparing the estimated position of the three-dimensional object and the traveling locus of the own vehicle and calculating a gap distance between the own vehicle and the three-dimensional object.

Images