JP3881048B2 - Narrow path guide device - Google Patents

Narrow path guide device Download PDF

Info

Publication number
JP3881048B2
JP3881048B2 JP28713995A JP28713995A JP3881048B2 JP 3881048 B2 JP3881048 B2 JP 3881048B2 JP 28713995 A JP28713995 A JP 28713995A JP 28713995 A JP28713995 A JP 28713995A JP 3881048 B2 JP3881048 B2 JP 3881048B2
Authority
JP
Japan
Prior art keywords
road
vehicle
narrow
obstacle
narrow road
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related
Application number
JP28713995A
Other languages
Japanese (ja)
Other versions
JPH09128687A (en
Inventor
敦 池田
多加志 木村
憲迩 上松
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Subaru Corp
Original Assignee
Fuji Jukogyo KK
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Fuji Jukogyo KK filed Critical Fuji Jukogyo KK
Priority to JP28713995A priority Critical patent/JP3881048B2/en
Publication of JPH09128687A publication Critical patent/JPH09128687A/en
Application granted granted Critical
Publication of JP3881048B2 publication Critical patent/JP3881048B2/en
Anticipated expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

Links

Images

Landscapes

  • Closed-Circuit Television Systems (AREA)
  • Traffic Control Systems (AREA)
  • Image Analysis (AREA)
  • Length Measuring Devices By Optical Means (AREA)
  • Control Of Driving Devices And Active Controlling Of Vehicle (AREA)
  • Processing Or Creating Images (AREA)

Description

【0001】
【産業上の利用分野】
本発明は、ADAシステム(アクティブ ドライブ アシスト システム)を搭載した自動車等の車両において、車両前方の狭路、その狭路の通行や障害物との接触の可能性を事前に判定して車両を安全にガイドする狭路ガイド装置に関する。
【0002】
【従来の技術】
近年、交通事故の増大傾向に対して車の安全性の飛躍的向上、即ち最初から危険な状態に陥らないように予防安全性を図るため、積極的に運転操作をアシストする総合的な運転支援システム、所謂ADAシステムが開発されている。このADAシステムは、車両に搭載した前方監視カメラにより撮像した画像信号により車両前方の風景から道路状況、障害物、交通環境等を認識して、この画像データにより車両の車線逸脱、追突、障害物との接触等の可能性を予測する。そして予測した場合の運転支援のあり方として、車はあくまで人間が操縦するという考えに基づき、先ず警報を発してドライバに回避操作を促す。警報を発してもドライバが適切に回避操作しない場合は、ブレーキ、スロットルまたはステアリングの運転操作系を安全側に自動的に制御したり、運転操作を一時的に代行するようにアシストすることが提案されている。
【0003】
上記ADAシステムとして、2台のカメラで撮像した画像信号を、ステレオ法による三角測量法で処理して距離を算出し、撮像した全領域にわたり距離分布を求めた三次元画像を作る方法が、既に本件出願人により提案されている。この方法によると、前方の撮像全域にわたり刻々と変化する道路形状、障害物等の距離データを作ることで、膨大な計算処理が必要になるが、撮像全域の三次元画像で道路形状、白線、道路上の障害物の位置や速度、障害物が何であるか等を高い精度で認識できる。このため車線逸脱、追突等の予測の正確性が向上するが、これ以外の有用な活用が期待されている。
【0004】
通常、狭い道路(狭路)を走行する場合は、ドライバが通行可能か否かの判断に迷うことがある。またドライバは運転席側の隙間は適切に把握できるが、反対の助手席側の隙間は把握しにくく、このため助手席側の障害物の通過をためらったり、無理に通過して障害物に接触することが間々ある。そこでこの狭路走行時に上記ADAシステムで車両を安全且つ適確にガイドすることが望まれる。
【0005】
従来、上記障害物等に対する車両のガイド装置に関しては、例えば特開平2−66352号公報の第1の先行技術があり、車体の側面に距離センサを設ける。そして交差点等の右左折の際に、距離センサによる車体とその側方の障害物との距離、操舵角により障害物と車体との接触の可能性を判定することが示されている。また特開昭61−6349号公報の第2の先行技術では、車両進行方向にレーザビームを発射し、その反射光を受光して、レーザ発射点に対する物標の方位、距離を算出する。そして算出結果、車速に基づいて物標との衝突の可能性、左右方向の通り抜け可能性、上下方向の通り抜け可能性を行うことが示されている。
【0006】
特開平6−293236号公報の第3の先行技術では、車両前方の風景を4つの領域に分割するために4個のレーザレーダを配置する。また4つの領域の画像信号を得るために、1個のCCDカメラ等のイメージセンサを設ける。そしてレーザレーダにより検出された方位で障害物が存在する領域を特定し、この領域に対応する二次元画像をイメージセンサにより得る。また二次元画像から障害物を抽出して、その障害物の大きさの指標を演算することが示されている。
【0007】
【発明が解決しようとする課題】
ところで、上記第1の先行技術にあっては、車両の右左折の際の車体側方の接触可能性に対処したものであるから、車両前方の障害物には対応できない。第2の先行技術にあっては、レーザを使用することで物標とそれまでの距離は算出できるが、物標が何であるかを識別したり、道路の形状や白線を検出することができないため、自車線の道路の障害物に対する接触の可能性等を正確に判断することが難しい。第3の先行技術にあっては、イメージセンサの二次元画像で特定された領域の障害物を検出するものであるから、狭路の通行可能性、この場合の障害物との接触可能性等を適確に判定できない等の問題がある。
【0008】
本発明は、このような点に鑑み、狭路の通行や障害物との接触可能性を適切に判定して、車両を安全にガイドすることを目的とする。
【0009】
【課題を解決するための手段】
この目的を達成するため、本発明の請求項1に係る狭路ガイド装置は、図1に示すように、前方監視カメラ11で撮像した画像信号で、車両前方の撮像された風景を距離分布の三次元画像データとして得る画像認識手段21と、この三次元画像の画像データにより前方の道路形状を認識すると共に、道路上の障害物を検出する道路、障害物検出手段22と、道路形状と障害物のデータによる道路の実質的な道幅と自車両の幅との関係により、事前に狭路か否かを判定する狭路判定手段31と、狭路判定手段31により狭路判定されたときに通行可能か否かを判定する通行判定手段32と、通行判定手段32により狭路の通行不可が判定されると第1の警報指令を発し、通行判定手段32により狭路の通行可能が判定されると第2の警報指令を発すると共に、通行判定手段32により狭路の通行可能が判定され、且つドライバの減速操作が判断された場合に、自車両前方における助手席側の一部域のみの三次元画像データを得るように画像認識手段21を切換えて、前方の撮像された風景の三次元画像データにより自車両の助手席側の障害物のみを認識し、この障害物との接触の可能性のみを監視する車両案内手段33とを備えることを特徴とする。
【0010】
請求項2に係る狭路ガイド装置は、請求項1に記載の発明の構成に加えて、前記車両案内手段33は、助手席側の障害物の現時点と数秒後との位置を比較し、自車両の助手席側と助手席側の障害物との接触の可能性を予測することを特徴とする。
【0012】
請求項3に係る狭路ガイド装置は、車両前方の狭路、その狭路の通行や障害物との接触の可能性を事前に判定して車両を安全にガイドする狭路ガイド装置において、自車両前方の道路幅を検出する検出手段と、前記検出手段が検出した道路幅と自車両の幅との関係により通行が可能な狭路か否かを判定する狭路判定手段と、前記狭路判定手段が通行可能な狭路と判定し、ドライバの減速操作が判断された場合に、前方のうち自車両の助手席側の障害物との接触の可能性のみを監視する車両案内手段とを備えることを特徴とする。
【0013】
請求項に係る狭路ガイド装置は、請求項に記載の発明の構成に加えて前記狭路判定手段は、道路形状、障害物のデータによる道路の実質的な道幅及び自車両の幅に基づいて狭路を判定することを特徴とする。
【0014】
【作用】
従って、本発明の請求項1にあっては、ドライバによる車両走行時に、前方の種々の立体物が前方監視カメラ11で撮像され、この画像信号により画像認識手段21で車両前方の撮像した風景に対する距離分布の三次元画像データが得られ、この画像データにより道路、障害物検出手段22で前方の道路形状を認識して道路上の障害物の位置、大きさ、速度等が検出される。また狭路判定手段31で道路形状と障害物のデータにより道路の実質的な道幅を求め、この道幅と車幅が比較され、このため道路が左右の障害物等により道幅が実質的に狭くなっている場合に、狭路であることが適切に判定される。
【0015】
そして通行判定手段32で狭路の通行不可が判定されると、車両案内手段33の第1警報指令により警報器13,14が警報を発し、このためドライバは通行不可が知らされて、無駄に進入しないように案内される。また、狭路の通行可能が判定されると、車両案内手段33の第2の警報指令により警報器13、14が警報を発し、このためドライバは通行可能であるがその通行には注意を要する狭路が前方に存在することを知らされると共に、前方の撮像された風景の三次元画像データより自車両の助手席側の障害物のみに対する接触可能性を集中して予測するので、処理する画像データが半分になった分、計算処理が少なくなると共に、処理する画像データの密度が倍になった分、助手席側の監視の精度が向上するので、助手席側の障害物に対する接触可能性がより早く、且つ正確に予測される。
【0016】
また、助手席側の障害物の現時点と数秒後との位置を比較することによって自車両の助手席側と助手席側の障害物との接触の可能性が予測される。これにより、自車両の把握しにくい助手席側が障害物に接触しないように監視された状態で安全に狭路を通行するように案内されるので、ドライバの負担が軽減する。
【0017】
また、狭路の通行可能が判定され、且つドライバの減速操作が判断された場合に、車両前方の一部域のみの三次元画像データが得られる。これにより、ドライバが狭路の通行に迷っている状況にあると助手席側が高い精度で監視されるので、狭路の通行が適切に行われるようになる。
【0018】
また、検出手段が検出した自車両前方の道路幅と自車両の幅との関係により通行が可能な狭路か否かが狭路判定手段31により判定される。さらに、この狭路判定手段31で通行可能な狭路と判定されると、車両案内手段33により自車両の助手席側の障害物との接触の可能性のみが監視される。これにより、助手席側の監視の精度が向上するので、助手席側の障害物に対する接触可能性がより早く、且つ正確に予測される。
【0019】
また、ドライバの減速操作が判断されると、車両案内手段33により自車両の助手席側の障害物との接触の可能性が監視される。これにより、ドライバが狭路の通行に迷っている状況にあると助手席側が高い精度で監視されるので、狭路の通行が適切に行われるようになる。また、狭路判定手段31により道路形状、障害物のデータによる道路の実質的な道幅及び自車両の幅に基づいて狭路が判定される。これにより、狭路が通行可能な狭路か否かが正確、且つ精度良く判定され、通行不可狭路への進入防止や進入による障害物との接触防止がさらに向上する。
【0020】
【実施例】
以下、本発明の実施例をADAシステムと関連づけた図面に基づいて説明する。図2において、車両とADAシステムの全体の概略について説明する。車両1は、車室前方の右側に運転席2が設けられ、運転席2の前にダッシュボード3、フロントウインド4、ハンドル5、インパネ6等が配置される。そこでドライバMが運転席に座り、フロントウインド4を介して前方の風景、道路、障害物等を見ながら運転操作することで、車両走行するようになっている。
【0021】
ADAシステム10は、車室内の天井7に前方監視カメラとして2台のCCDカメラ11が装着される。2台のCCDカメラ11は、左右に離間して配置され、車両前方の風景をステレオ式に撮像するのであり、このカメラ11の画像信号が制御ユニット12に入力して処理される。また警報手段として、例えば車室内に2個の警報器13,14が設けられている。警報器13,14は、例えば警報の光と音を併用するものとして、警報ランプ15とスピーカ16とが一体形成される。そしてこの警報器13,14が、運転席2のドライバMを中心としてその前方の左右両側で視界内の視界の妨げにならない場所、例えばフロントウインド4とダッシュボード3の境目付近にそれぞれ配置されている。
【0022】
警報ランプ15は、ドライバMの目の位置、所謂アイポイントに所定の色の光Hを絞り且つ点滅して照射して、日中に光量が少なくても明確に視認することが可能になっている。これにより光の漏洩を抑えて、同乗者や他の車両への影響を低減することも可能になる。スピーカ16は、所定の音量の例えば断続した音Oを発生するものである。そして制御ユニット12からの警報信号が左右警報器13,14の一方または両方に出力して、ドライバMを中心としてその左右の一方または両方から警報を発するように構成される。
【0023】
図3において、制御ユニットの機能ブロック図について説明する。制御ユニット12は、2台のCCDカメラ11の画像信号により車線逸脱、追突、障害物接触等の種々の可能性を予測するが、本発明に関係するものとして障害物接触警報システム20と狭路ガイドシステム30を備える。障害物接触警報システム20は、2台のCCDカメラ11の画像信号が入力する全画面の画像認識手段21を有する。画像認識手段21は、車両前方の撮像された風景を距離分布の三次元画像データとして物体を認識するものであり、撮像した風景を小領域に分割してその各々について、物体をステレオ画像処理して三次元的な位置を算出し、各物体の三次元座標の距離分布の画像データを得る。そしてこの画像データが道路、障害物検出手段22に入力して、前方の道路形状を認識し、同時に道路上の障害物を検出する。
【0024】
即ち、個々の物体の三次元距離データにより、物体が重なっている場合にも容易に分離して、何であるかを認識することができ、これにより白線、ガードレール、縁石等を抽出する。そして例えば白線を折れ線で近似して、左右の折れ線で囲まれた範囲を自車線と判断し、この自車線のデータの水平成分で道路のカーブを、垂直成分で道路の上りや下りを検出して、手前から遠方に向かって道路形状を三次元的に検出する。また画像データを道路形状と比較して道路上の障害となる立体物のみを選別し、この選別された物体が自車線の障害物であるのかを判断すると共に、その障害物までの距離、大きさ、左右端の位置を検出するのであり、この道路形状と障害物のデータが接触判定手段23に入力する。
【0025】
接触判定手段23は、障害物の左右端の位置とその時間変化から数秒後の位置を推定し、道路形状から得られる自車両の数秒後の位置を推定し、両者を比較して障害物が自車両の車体の左右のいずれかに接触するか否かの可能性を予測する。そして左右の接触可能性を予測するとその信号が警報制御手段24に入力し、左側接触の場合は左警報器13に警報信号を出力して警報を発する。また右側接触の可能性有りの場合は、右警報器14に警報信号を出力して警報を発する。
【0026】
そこでドライバMが例えば居眠り等で覚醒度が低下したり、他に気をとられて運転することで自車両の左右の一方に前方の障害物が接触する可能性を生じると、ドライバMを中心として接触可能性有りの側の警報器13または14が警報を発する。従って、ドライバMはその警報で直ちに接触の可能性と、接触可能性有りの側とを同時に認知することができ、このため迷うことなく減速したり、接触可能性有りの側と反対にハンドル操作して、接触回避することが可能となる。
【0027】
ここで上記障害物接触警報システム20は、車体が障害物に極接近して数秒後に接触する可能性がある場合に警報するものである。従って、障害物等により道路の道幅が実質的に狭くなっている狭路での車両ガイド機能を備えていない。このため狭路等の道路状況では、各別の車両ガイドシステムを設けることが望まれる。
【0028】
そこで本発明においては、後述する狭路ガイドシステム30を設け、狭路のガイドに役立てようとするものである。狭路ガイドシステム30は、三つの要素により構成され、第1構成要素である道路形状と障害物のデータが入力する狭路判定手段31は、自車線で極低速または静止している車両、道路端部のガードレール、縁石、家屋の塀等の障害物の間隔を計測して、道路の実質的な道幅Dを検出する。そして道幅Dと自車両の車体の最大幅W及び余裕分との関係で、例えば道幅Dが車体の最大幅Wに40cmの余裕分を加算した値より小さい場合に、狭路を判定する。この狭路の信号は、第2構成要素である通行判定手段32に入力し、道幅Dが車体の最大幅Wに余裕分として、例えば10cm(最小値),40cm(最大値)を加算した値と比較する。ここで、道幅Dが車体の最大幅Wに余裕分(最小値)を加えた値より小さい場合に通行不可を判定し、W+10<D<W+40の場合にぎりぎりの通行可能を判定する。
【0029】
この通行判定結果の信号は第2構成要素である車両案内手段33に入力する。車両案内手段33は、前記通行判定手段32の判定結果を出力するタイミングを決定して信号を発する距離判定手段34と、この距離判定手段34により決定されたタイミングに同期して前記車両前方全域の三次元距離分布画像データを車両前方一部域のみの三次元距離分布画像を得るように画像処理計算を切換える切換手段36とを有する。
【0030】
距離判定手段34は、通行判定手段32により通行不可が判定されるとき、第1の信号を警報制御手段24へ出力し、警報制御手段24は左右警報器13,14を第1態様(例えば断続音)で作動させる。そして、自車両が、自車両の車速Vと予め設定した減速度α(例えば0.5G)から演算で求まる制動距離(V2/2α)に予め設定した作動遅れ時間Tによる空走距離(V・T)を加算した距離LSに達すると、第3の信号を警報制御手段24へ出力し警報制御手段24は左右警報器13,14を第3態様(例えば連続音)で作動させる。
【0031】
一方、距離判定手段34は、通行判定手段32により通行可能が判定されるとき第2の信号を警報制御手段24へ出力し、警報制御手段24は左右警報器13,14を第2態様(例えば長短音)で作動させる。そして、ドライバによる減速操作を検知すると計算処理切換手段36へ切換指令信号を出力し、計算処理切換手段36の切換信号により画像認識手段21は全域三次元画像を一部域三次元画像を得るように切換わると共に、警報制御手段24は運転席側警報器への通電回路を開路する。
【0032】
尚、ドライバによる減速操作の検知は、アクセルスイッチ26、ブレーキスイッチ27、加速度センサ28からの信号により検知される。
【0033】
次に、狭路走行での警報制御を、図4のフローチャートにより説明する。車両走行時に、CCDカメラ11による画像信号で車両前方の撮像された風景を認識し(ステップS1)、道路形状と道路上の障害物を検出する(ステップS2)。そして道幅Dを車体の最大幅Wに余裕分(最大値)として40cmを加算した値と比較し(ステップS3)、D≧W+40の場合は余裕の多い通常の道路を判断してステップ1に戻る。一方、D<W+40の場合は、通行可能な狭路を判定し(ステップS4)、狭路ガイドシステム30が作動する。
【0034】
そこで例えば図5(a)のように自車両Aの前方の道路Bの右側が縁石Eで左側に車両Cが駐車している状況、あるいは図5(b)のように自車両Aの前方の道路Bの左右に、前後に離間して車両C,Cが駐車している状況において、車両と縁石あるいは車両C,C同士の間隔により道路Bの実質的な道幅Dを計測して、その道路Bが余裕の少ない場合に狭路であることが、事前に判定される。更に道幅Dを車体の最大幅Wに余裕分(最小値)として例えば10cmを加算した値と比較し(ステップS5)、D<W+10の場合は通行不可を判定し(ステップS6)、第1態様(例えば断続音)で警報する(ステップ7)。このため道幅Dに余裕が殆ど無い場合は、狭路が通行不可であることが、進入前に適切に判定される。そして、前記警報によりドライバは狭路侵入前に前方の狭路が通行できない狭路であることを知らされ、無駄に進入して立往生することがないよう案内される。
【0035】
前記警報された後、自車両の車速Vと予め設定した減速度α(例えば0.5G)から演算で求まる制動距離(V/V・T)に予め設定した作動遅れ時間Tによる空走距離(V・T)を加算した距離LSを算出し(ステップS8)、更に通行不可と判定した狭路への接近を続けるときは、自車両と通行不可と判定した狭路までの距離Lと前記LSが等しくなったか否かを判断し(ステップS9)、その時点で第3態様の警報を発する(ステップS10)。この警報によりドライバは制動操作の必要を知らされ、直ちに制動操作して停車することで通行不可狭路への進入防止や進入による障害物との接触防止を図ることができる。
【0036】
一方、図5(c)のように道路の道幅に少し余裕があって、W+10<D<W+40の場合は、ぎりぎり通行可能であることが判定される。この場合は第2態様(例えば長短音)の警報を発し(ステップS11)、これによりドライバの注意が促される。そして、次のステップに移行すると、アクセルスイッチ26、ブレーキスイッチ27または加速度センサ28の信号によりドライバによる減速操作があったか否かをチェックし(ステップS12)、減速操作が無い場合は、ドライバが狭路の通行に迷いが無いことを判断してそのまま抜ける。そこで無駄な計算処理が省略される。この場合は、通常の障害物接触警報システム20で車体の左右の接触可能性が予測される。これに対し、減速操作が有った場合には、計算処理切換手段36が一部域(助手席側)のみ計算処理するように切換えられる(ステップ13)。
【0037】
すなわち、減速操作が有った場合は、ドライバが狭路の通行に戸惑っていると判断し、一部域(助手席側)のみ計算処理するように切換える(ステップS13)。そこで図5(d)のように前方の撮像された全域Fにおいて助手席側半分の障害物C’側だけを認識し、同図(e)のようにこの障害物Cに対して自車両Aの助手席側部A’の現時点と数秒後の位置を比較し、接触可能性を予測する。このとき、処理すべき画像データが半分になるが、その分画像データの密度を倍にし、精度の向上を図る。または、画像データが減った分、計算処理を早く行えることができるので、より早い予測が可能になる。従って、ドライバが通行の判断に迷っている状況において、自車両Aの助手席側が狭路ガイドシステム30により高い精度あるいはより早い予測で監視されて、適切に案内される。
【0038】
そこでドライバが狭路を注意深く通行する際に、自車両Aの左側の障害物Cに対する接触可能性を判断し(ステップS14)、自車両Aが左に寄って障害物Cとの接触可能性を生じると、直ちに左警報器13で警報を発する(ステップS15)。従って、ドライバはこの警報で車体左側の接触可能性が事前に知らされ、このため右側にハンドル操作して回避行動を行うことで、接触防止される。
【0039】
こうして自車両Aの助手席側が高い精度あるいはより早い予測で監視されることで、ドライバは主として把握し易い運転席側を監視して通行すれば良くなって、ドライバの負担が軽減する。そして自車両A自体も、左右が狭路ガイドシステム30とドライバにより監視されて、狭路を安全に通過することが可能になる。そして狭路を脱すると、通常の道路を判断して通常のシステムに戻る。
【0040】
以上、本発明の実施例について説明したが、左ハンドルの場合は狭路の通行可能の場合に、助手席の右側の障害物に対する接触可能性を予測することは勿論である。狭路の判定方法、狭路で通行可能な場合の接触可能性の予測は、実施例に限定されない。一部画面で画像認識したものは、車両の左側の接触可能性以外にも適応できる。
【0041】
【発明の効果】
以上に説明したように、本発明に係る狭路ガイド装置では、前方監視カメラで撮像した画像信号で、車両前方の撮像された風景を距離分布の三次元画像データとして得る画像認識手段と、この三次元画像の画像データにより前方の道路形状を認識すると共に、道路上の障害物を検出する道路、障害物検出手段と、道路形状と障害物のデータによる道路の実質的な道幅と自車両の幅との関係により、事前に狭路か否かを判定する狭路判定手段と、狭路判定されたときに通行可能か否かを判定する通行判定手段と、狭路の通行不可が判定されると第1の警報指令を発し、狭路の通行可能が判定されると第2の警報指令を発し自車両が障害物と接触しないように監視すると共に、前方の撮像された風景の三次元画像データにより自車両の助手席側の障害物のみを認識し、この障害物との接触の可能性のみを監視する車両案内手段とを備える構成である。これにより、道幅の狭い狭路を事前に精度良く判定できるので、狭路が通行不可の場合は進入前の警報により、また、通行可能な場合は自車両の助手席側の障害物のみに対する接触可能性を集中して予測、監視するので、処理する画像データが半分になった分、計算処理が少なくなると共に、処理する画像データの密度が倍になった分、助手席側の監視の精度が向上するので、助手席側の障害物に対する接触可能性をより早く、且つ正確に予測することができるようになる。
【0042】
また、助手席側の障害物の現時点と数秒後との位置を比較することによって自車両の助手席側と助手席側の障害物との接触の可能性が予測される。これにより、自車両の把握しにくい助手席側が障害物に接触しないように監視された状態で安全に狭路を通行するように案内されるので、ドライバの負担を軽減することができる。
【0043】
また、狭路の通行可能が判定され、且つドライバの減速操作が判断された場合に、車両前方の一部域のみの三次元画像データが得られる。これにより、ドライバが狭路の通行に迷っている状況にあると助手席側が高い精度で監視されるので、狭路の通行をより適切に行うことができるようになる。
【0044】
また、検出手段が検出した自車両前方の道路幅と自車両の幅との関係により通行が可能な狭路か否かが狭路判定手段により判定される。さらに、この狭路判定手段で通行可能な狭路と判定されると、車両案内手段により自車両の助手席側の障害物との接触の可能性のみが監視される。これにより、助手席側の監視の精度が向上するので、助手席側の障害物に対する接触可能性がより早く、且つ正確に予測できるようになる。
【0045】
また、ドライバの減速操作が判断されると、車両案内手段により自車両の助手席側の障害物との接触の可能性が監視される。これにより、ドライバが狭路の通行に迷っている状況にあると助手席側が高い精度で監視されるので、狭路の通行を適切に行うことができるようになる。
また、狭路判定手段により道路形状、障害物のデータによる道路の実質的な道幅及び自車両の幅に基づいて狭路が判定される。これにより、狭路が通行可能な狭路か否かが正確、且つ精度良く判定され、通行不可狭路への進入防止や進入による障害物との接触防止をさらに向上させることができる。
【図面の簡単な説明】
【図1】本発明に係る狭路ガイド装置の構成を示すクレーム対応図である。
【図2】ADAシステムの概略を示す説明図である。
【図3】本発明の実施例の機能ブロック図である。
【図4】狭路の警報制御のフローチャートである。
【図5】狭路通行の作動状態を示す説明図である。
【符号の説明】
11 CCDカメラ(前方監視カメラ)
13,14 警報器
21 画像認識手段
22 道路、障害物検出手段
31 狭路判定手段
32 通行判定手段
33 車両案内手段
[0001]
[Industrial application fields]
In the present invention, in vehicles such as automobiles equipped with an ADA system (active drive assist system), it is possible to make a vehicle safe by determining in advance the possibility of narrow roads in front of the vehicle, passage of the narrow roads, and contact with obstacles. The present invention relates to a narrow path guide device that guides to a vehicle.
[0002]
[Prior art]
Comprehensive driving assistance that actively assists driving operations in recent years to dramatically improve car safety against the increasing trend of traffic accidents, that is, preventive safety from falling into a dangerous state from the beginning Systems, so-called ADA systems, have been developed. This ADA system recognizes road conditions, obstacles, traffic environments, and the like from the scenery in front of the vehicle based on an image signal captured by a front monitoring camera mounted on the vehicle, and uses this image data to deviate from the lane of the vehicle, a rear-end collision, and an obstacle. Predict the possibility of contact with Based on the idea that the vehicle is maneuvered by humans as a way of driving assistance when predicted, first, an alarm is issued to prompt the driver to perform an avoidance operation. If the driver does not properly avoid the operation even if an alarm is issued, it is suggested to automatically control the driving operation system of the brake, throttle or steering to the safe side, or assist the driver to temporarily substitute the driving operation Has been.
[0003]
As the ADA system, there is already a method of calculating a distance by processing image signals captured by two cameras by a triangulation method using a stereo method, and creating a three-dimensional image in which a distance distribution is obtained over the entire captured region. Proposed by the applicant. According to this method, it is necessary to make a huge amount of calculation processing by creating distance data such as road shapes and obstacles that change every moment over the entire imaging area in front, but the road shape, white line, The position and speed of the obstacle on the road, what the obstacle is, and the like can be recognized with high accuracy. For this reason, the accuracy of predictions such as lane departure and rear-end collision is improved, but other useful uses are expected.
[0004]
Usually, when driving on a narrow road (narrow road), the driver may be at a loss as to whether or not the vehicle can pass. The driver can properly grasp the gap on the driver's seat side, but it is difficult to grasp the gap on the opposite passenger's side, so hesitates to pass obstacles on the passenger's side or forcibly passes to touch the obstacles There is a lot to do. Therefore, it is desired to guide the vehicle safely and appropriately with the ADA system when traveling on this narrow road.
[0005]
Conventionally, there is a first prior art disclosed in, for example, Japanese Patent Application Laid-Open No. 2-66352 regarding a vehicle guide device for the obstacle and the like, and a distance sensor is provided on a side surface of the vehicle body. In the case of a right or left turn at an intersection or the like, it is shown that the possibility of contact between the obstacle and the vehicle body is determined based on the distance and the steering angle between the vehicle body and the obstacle on its side by the distance sensor. In the second prior art disclosed in Japanese Patent Application Laid-Open No. 61-6349, a laser beam is emitted in the vehicle traveling direction, the reflected light is received, and the azimuth and distance of the target with respect to the laser emission point are calculated. The calculation result shows that the possibility of collision with the target, the possibility of passing through in the left-right direction, and the possibility of passing through in the up-down direction are performed based on the vehicle speed.
[0006]
In the third prior art disclosed in Japanese Patent Laid-Open No. Hei 6-293236, four laser radars are arranged to divide the landscape in front of the vehicle into four regions. Further, an image sensor such as one CCD camera is provided in order to obtain image signals of four regions. Then, an area where an obstacle is present is specified in the direction detected by the laser radar, and a two-dimensional image corresponding to this area is obtained by an image sensor. Further, it is shown that an obstacle is extracted from a two-dimensional image and an index of the obstacle size is calculated.
[0007]
[Problems to be solved by the invention]
By the way, in the said 1st prior art, since it copes with the contact possibility of the vehicle body side at the time of the right and left turn of a vehicle, it cannot respond to the obstruction ahead of a vehicle. In the second prior art, it is possible to calculate the target and the distance to it by using a laser, but it is impossible to identify what the target is and to detect the shape of the road and the white line. Therefore, it is difficult to accurately determine the possibility of contact with obstacles on the road in the own lane. In the third prior art, an obstacle in the area specified by the two-dimensional image of the image sensor is detected. Therefore, the possibility of passage through a narrow path, the possibility of contact with the obstacle in this case, etc. There is a problem that it cannot be determined accurately.
[0008]
In view of such a point, an object of the present invention is to appropriately determine the possibility of passage through a narrow path or contact with an obstacle and guide the vehicle safely.
[0009]
[Means for Solving the Problems]
  In order to achieve this object, a narrow road guide device according to claim 1 of the present invention is a distance distribution of a landscape imaged in front of a vehicle using an image signal imaged by a front monitoring camera 11, as shown in FIG. Image recognition means 21 obtained as three-dimensional image data, a road for detecting obstacles on the road, obstacle detection means 22, road shape and obstacles, while recognizing the road shape ahead by the image data of the three-dimensional image Narrow road determination means 31 for determining in advance whether or not the road is a narrow road according to the relationship between the substantial road width of the road and the width of the host vehicle based on the object dataBy narrow path judging means 31Passage determination means 32 for determining whether or not the passage is possible when the narrow path is determined;By the traffic judging means 32When it is determined that a narrow road is not permitted, a first warning command is issued.By the traffic judging means 32When it is determined that the narrow road is accessible, a second warning command is issued,Image recognition means so as to obtain three-dimensional image data of only a partial area on the passenger seat side in front of the host vehicle when it is determined that the passage determination means 32 allows passage of a narrow road and the driver's deceleration operation is determined. 21 is switched,Vehicle guidance means 33 for recognizing only the obstacle on the passenger seat side of the own vehicle from the captured three-dimensional image data of the scenery in front and monitoring only the possibility of contact with the obstacle is provided.GetIt is characterized by that.
[0010]
  The narrow path guide device according to claim 2 is:In addition to the configuration of the invention described in claim 1, the vehicle guide means 33 compares the position of the obstacle on the passenger seat side with the current position and a few seconds later, and the obstacle on the passenger seat side and the passenger seat side of the own vehicle Predict the possibility of contact with thingsIt is characterized by that.
[0012]
  The narrow road guide device according to claim 3 is a narrow road guide device that safely guides the vehicle by determining in advance the narrow road ahead of the vehicle, passage of the narrow road, and the possibility of contact with an obstacle. Detection means for detecting a road width in front of the vehicle, narrow road determination means for determining whether or not the road is capable of passing according to a relationship between the road width detected by the detection means and the width of the host vehicle, and the narrow road Judged as a narrow path that can be passed by the judging meansWhen the driver's deceleration operation is determined,Vehicle guidance means for monitoring only the possibility of contact with obstacles on the passenger seat side of the vehicle ahead of the vehicleGetIt is characterized by that.
[0013]
  Claim4The narrow path guide device according to claim3In addition to the configuration of the invention described in,The narrow road determining means determines the narrow road based on a road shape, a substantial road width of the road based on obstacle data, and a width of the host vehicle.
[0014]
[Action]
Therefore, according to the first aspect of the present invention, when the vehicle is driven by the driver, various three-dimensional objects ahead are imaged by the front monitoring camera 11, and the image recognition means 21 is used for the image captured in front of the vehicle by this image signal. Three-dimensional image data of the distance distribution is obtained, and the position, size, speed, etc. of the obstacle on the road are detected by recognizing the road shape ahead by the road / obstacle detection means 22 from this image data. Further, the narrow road judging means 31 obtains the substantial road width of the road from the road shape and the obstacle data, and the road width and the vehicle width are compared. For this reason, the road is substantially narrowed by the left and right obstacles. If it is, it is appropriately determined that the road is narrow.
[0015]
  When the passage determination unit 32 determines that the passage of the narrow road is impossible, the alarm devices 13 and 14 issue an alarm according to the first warning command of the vehicle guide unit 33, and thus the driver is informed that the passage is impossible and is useless. You will be guided not to enter. Further, when it is determined that the passage through the narrow road is possible, the alarm devices 13 and 14 issue an alarm in response to the second warning command of the vehicle guide means 33, so that the driver can pass but requires caution. Informed that the narrow streets are ahead,Since the possibility of contact with only the obstacle on the passenger seat side of the vehicle is predicted in a concentrated manner from the three-dimensional image data of the scenery taken in front of the vehicle, the calculation processing is reduced because the image data to be processed is halved. At the same time, since the density of the image data to be processed is doubled, the monitoring accuracy on the passenger seat side is improved, so that the possibility of contact with the obstacle on the passenger seat side is predicted more quickly and accurately.
[0016]
Also,By comparing the current position of the obstacle on the passenger seat side and the position after a few seconds, the possibility of contact between the obstacle on the passenger seat side and the passenger seat side of the host vehicle is predicted. As a result, the passenger's side of the vehicle, which is difficult to grasp, is guided to safely pass through the narrow road in a state where it is monitored so as not to touch an obstacle, so that the burden on the driver is reduced.
[0017]
Also,When it is determined that the narrow road can be passed and the driver's deceleration operation is determined, three-dimensional image data of only a partial area in front of the vehicle is obtained. As a result, if the driver is in a situation where he / she is confused about the passage of the narrow road, the passenger seat side is monitored with high accuracy, so that the passage of the narrow road is appropriately performed.
[0018]
Also,The narrow road determination means 31 determines whether or not the road is a narrow road that can be passed by the relationship between the road width in front of the host vehicle detected by the detection means and the width of the host vehicle. Further, if the narrow road determining means 31 determines that the road is passable, the vehicle guide means 33 monitors only the possibility of contact with the obstacle on the passenger seat side of the own vehicle. Thereby, since the accuracy of monitoring on the passenger seat side is improved, the possibility of contact with the obstacle on the passenger seat side is predicted more quickly and accurately.
[0019]
Also,When the driver's deceleration operation is determined, the vehicle guide means 33 monitors the possibility of contact with an obstacle on the passenger seat side of the host vehicle. As a result, if the driver is in a situation where he / she is confused about the passage of the narrow road, the passenger seat side is monitored with high accuracy, so that the passage of the narrow road is appropriately performed.Also,The narrow road determination means 31 determines a narrow road based on the road shape, the substantial road width of the road based on the obstacle data, and the width of the host vehicle. Accordingly, it is accurately and accurately determined whether or not the narrow road is passable, and further prevention of entry into the inaccessible narrow road and contact with an obstacle due to the entry is further improved.
[0020]
【Example】
Embodiments of the present invention will be described below with reference to the drawings associated with an ADA system. In FIG. 2, the outline of the entire vehicle and the ADA system will be described. In the vehicle 1, a driver's seat 2 is provided on the right side in front of the passenger compartment, and a dashboard 3, a front window 4, a handle 5, an instrument panel 6, and the like are disposed in front of the driver's seat 2. Therefore, the driver M sits in the driver's seat and drives the vehicle by driving the vehicle while looking at the scenery, roads, obstacles, and the like ahead through the front window 4.
[0021]
In the ADA system 10, two CCD cameras 11 are mounted as a front monitoring camera on a ceiling 7 in a vehicle interior. The two CCD cameras 11 are arranged so as to be separated from each other on the left and right, and take a stereo image of the scenery in front of the vehicle. The image signals of the cameras 11 are input to the control unit 12 and processed. As alarm means, for example, two alarm devices 13 and 14 are provided in the passenger compartment. In the alarm devices 13 and 14, for example, an alarm lamp 15 and a speaker 16 are integrally formed as a combination of alarm light and sound. The alarms 13 and 14 are arranged at locations where the visual field in the field of view is not obstructed on the left and right sides of the driver M of the driver's seat 2, for example, near the boundary between the front window 4 and the dashboard 3. Yes.
[0022]
The warning lamp 15 irradiates and blinks light H of a predetermined color on the eyes of the driver M, that is, so-called eyepoints, so that it can be clearly seen even during the day when the amount of light is small. Yes. As a result, light leakage can be suppressed and the influence on passengers and other vehicles can be reduced. The speaker 16 generates, for example, an intermittent sound O having a predetermined volume. An alarm signal from the control unit 12 is output to one or both of the left and right alarm devices 13 and 14, and the driver M is configured to issue an alarm from one or both of the left and right.
[0023]
In FIG. 3, a functional block diagram of the control unit will be described. The control unit 12 predicts various possibilities such as lane departure, rear-end collision, obstacle contact, and the like based on the image signals of the two CCD cameras 11, but the obstacle contact alarm system 20 and the narrow path are related to the present invention. A guide system 30 is provided. The obstacle contact alarm system 20 has a full-screen image recognition means 21 to which image signals from two CCD cameras 11 are input. The image recognizing means 21 recognizes an object using the captured landscape in front of the vehicle as the three-dimensional image data of the distance distribution, divides the captured landscape into small regions, and performs stereo image processing on the object for each of them. The three-dimensional position is calculated to obtain image data of the distance distribution of the three-dimensional coordinates of each object. This image data is input to the road / obstacle detection means 22 to recognize the front road shape and simultaneously detect the obstacle on the road.
[0024]
In other words, the three-dimensional distance data of individual objects can be easily separated even when the objects are overlapped to recognize what they are, thereby extracting white lines, guardrails, curbs, and the like. For example, the white line is approximated by a broken line, the range enclosed by the left and right broken lines is determined as the own lane, the road curve is detected with the horizontal component of this lane data, and the up and down of the road is detected with the vertical component. Then, the road shape is detected three-dimensionally from the front to the far side. In addition, the image data is compared with the road shape to select only three-dimensional obstacles on the road, determine whether the selected object is an obstacle in the own lane, and the distance and size to the obstacle. The left and right end positions are detected, and the road shape and obstacle data are input to the contact determination means 23.
[0025]
The contact determination means 23 estimates the position after a few seconds from the position of the left and right ends of the obstacle and its change over time, estimates the position after a few seconds of the host vehicle obtained from the road shape, and compares the two to compare the obstacle Predict the possibility of contact with either the left or right side of the vehicle body. When the left / right contact possibility is predicted, the signal is input to the alarm control means 24. When the left contact is detected, an alarm signal is output to the left alarm device 13 to issue an alarm. If there is a possibility of right-side contact, an alarm signal is output to the right alarm device 14 to issue an alarm.
[0026]
Therefore, if the driver M has a possibility that the awakening level is reduced due to, for example, a snooze or the like, or if the driver M is driving with other attention, the obstacle in front may come into contact with one of the left and right sides of the own vehicle. As a result, the alarm device 13 or 14 on the side with the possibility of contact issues an alarm. Therefore, the driver M can simultaneously recognize the possibility of contact and the side with the possibility of contact by the alarm, so that the driver M can decelerate without hesitation or operate the handle opposite to the side with the possibility of contact. Thus, it is possible to avoid contact.
[0027]
Here, the obstacle contact warning system 20 warns when there is a possibility that the vehicle body may come close to the obstacle and come into contact several seconds later. Therefore, the vehicle guide function is not provided in a narrow road where the road width is substantially narrow due to an obstacle or the like. For this reason, it is desirable to provide a separate vehicle guide system in road conditions such as narrow roads.
[0028]
Therefore, in the present invention, a narrow path guide system 30 to be described later is provided so as to be useful for narrow path guidance. The narrow road guide system 30 is composed of three elements, and the narrow road determination means 31 for inputting road shape and obstacle data, which is the first constituent element, is a vehicle or road that is extremely slow or stationary in its own lane. By measuring the distance between obstacles such as guardrails at the end, curbs, and fences of houses, the substantial road width D of the road is detected. A narrow road is determined when the road width D is smaller than a value obtained by adding a margin of 40 cm to the maximum width W of the vehicle body because of the relationship between the road width D and the maximum width W of the vehicle body and the margin. The narrow road signal is input to the traffic determination means 32 as the second component, and the road width D is a value obtained by adding, for example, 10 cm (minimum value) and 40 cm (maximum value) as a margin to the maximum width W of the vehicle body. Compare with Here, when the road width D is smaller than a value obtained by adding a margin (minimum value) to the maximum width W of the vehicle body, it is determined that the vehicle cannot pass, and when W + 10 <D <W + 40, it is determined that the vehicle can pass only.
[0029]
This traffic determination result signal is input to the vehicle guide means 33 which is the second component. The vehicle guide means 33 determines the timing for outputting the determination result of the traffic determination means 32 and generates a signal, and the distance determination means 34 generates a signal, and synchronizes with the timing determined by the distance determination means 34 in the entire area in front of the vehicle. And switching means 36 for switching the image processing calculation so as to obtain the three-dimensional distance distribution image data of the three-dimensional distance distribution image of only a partial area in front of the vehicle.
[0030]
  The distance determination unit 34 outputs a first signal to the alarm control unit 24 when the passage determination unit 32 determines that the passage is impossible.13, 14Is operated in the first mode (for example, intermittent sound). Then, the vehicle travels in the idle travel distance (V · V) by the preset operation delay time T to the braking distance (V2 / 2α) obtained by calculation from the vehicle speed V of the host vehicle and a preset deceleration α (for example, 0.5 G). When the distance LS obtained by adding T) is reached, the third signal is output to the alarm control means 24, and the alarm control means 2413, 14Is operated in a third manner (for example, continuous sound).
[0031]
  On the other hand, the distance determination means 34 outputs a second signal to the alarm control means 24 when the passage determination means 32 determines that the passage is possible.13, 14Is operated in the second mode (for example, long and short sounds). And driverMWhen the decelerating operation is detected, a switching command signal is output to the calculation processing switching means 36, and the image recognition means 21 is switched by the switching signal of the calculation processing switching means 36 so as to obtain a partial three-dimensional image. At the same time, the alarm control means 24 opens an energization circuit to the driver's seat side alarm device.
[0032]
  DriverMThe deceleration operation is detected by signals from the accelerator switch 26, the brake switch 27, and the acceleration sensor 28.
[0033]
Next, alarm control in narrow road traveling will be described with reference to the flowchart of FIG. When the vehicle travels, the imaged landscape in front of the vehicle is recognized by the image signal from the CCD camera 11 (step S1), and the road shape and obstacles on the road are detected (step S2). Then, the road width D is compared with a value obtained by adding 40 cm to the maximum width W of the vehicle body as a margin (maximum value) (step S3). If D ≧ W + 40, a normal road with a large margin is determined and the process returns to step 1. . On the other hand, when D <W + 40, a narrow path that can be passed is determined (step S4), and the narrow path guide system 30 is activated.
[0034]
  Therefore, for example, the situation where the right side of the road B ahead of the host vehicle A is curb E and the vehicle C is parked on the left side as shown in FIG. 5A, or the front side of the host vehicle A as shown in FIG. In a situation where the vehicles C and C are parked on the left and right of the road B, separated from each otherCAnd curbEOr vehicleC, CA substantial road width D of the road B is measured based on an interval between the roads, and when the road B has a small margin, it is determined in advance that the road is a narrow road. Further, the road width D is compared with a value obtained by adding, for example, 10 cm as a margin (minimum value) to the maximum width W of the vehicle body (step S5). If D <W + 10, it is determined that the vehicle cannot pass (step S6). An alarm is given (for example, intermittent sound) (step 7). For this reason, when there is almost no margin in the road width D, it is appropriately determined that the narrow road is not allowed to pass before entering. And the driver by the alarmMIs informed that the narrow road ahead cannot pass through before entering the narrow road, and is guided so as not to get stuck and get stuck.
[0035]
  After being alerted, the vehicleABraking distance (V) obtained by calculation from a vehicle speed V of the vehicle and a preset deceleration α (for example, 0.5 G)2When the distance LS is calculated by adding the idling distance (V · T) based on the preset operation delay time T to (V · T) (step S8), and further when approaching the narrow road determined to be impassable , Own vehicleAIt is determined whether or not the distance L to the narrow road determined to be impassable and the LS are equal (step S9), and a third mode alarm is issued at that time (step S10). The driver by this alarmMIs informed of the necessity of the braking operation, and can immediately prevent the vehicle from entering the inaccessible narrow road or contact with the obstacle due to the approach by stopping the vehicle by braking.
[0036]
  On the other hand, as shown in FIG. 5 (c), when there is a margin in the road width and W + 10 <D <W + 40, it is determined that the vehicle can pass through. In this case, a warning of the second mode (for example, long and short sounds) is issued (step S11), thereby the driverMThe attention of is urged. And, Going to the next step,A driver based on a signal from an accelerator switch 26, a brake switch 27, or an acceleration sensor 28MIt is checked whether or not there has been a deceleration operation by (step S12), and if there is no deceleration operation, the driverMJudges that there is no hesitation in the passage of narrow streets and goes out as it is. Therefore, useless calculation processing is omitted. In this case, the possibility of contact on the left and right of the vehicle body is predicted by the normal obstacle contact alarm system 20.On the other hand, when there is a deceleration operation, the calculation processing switching means 36 is switched so as to perform calculation processing only in a partial area (passenger seat side) (step 13).
[0037]
  That is,If there is a deceleration operation, the driverMFor traffic on narrow streetsConfusedSwitching to perform calculation processing only on a partial area (passenger side) (step S13). Therefore, as shown in FIG. 5D, only the obstacle C ′ side of the passenger seat side half is recognized in the entire imaged area F in the front, and the own vehicle A against the obstacle C as shown in FIG. The current position of the passenger seat side part A ′ is compared with the position after a few seconds to predict the possibility of contact. At this time, the image data to be processed is halved, but the image data density is doubled accordingly,accuracyTo improve. Alternatively, since the calculation process can be performed earlier as much as the image data is reduced, faster prediction is possible. Therefore, the driverMHowever, the passenger seat side of the host vehicle A is monitored with high accuracy or faster prediction by the narrow road guide system 30 and is guided appropriately.
[0038]
  Driver thereMWhen the vehicle carefully passes through the narrow road, the possibility of contact with the obstacle C on the left side of the own vehicle A is judged (step S14), and when the own vehicle A approaches the left and the contact possibility with the obstacle C occurs. Immediately, the left alarm device 13 issues an alarm (step S15). Therefore, the driverMThe possibility of contact on the left side of the car body with this alarmBeforehandFor this reason, contact is prevented by operating the steering wheel on the right side and performing an avoidance action.
[0039]
  Thus, the driver's seat side of the vehicle A is monitored with high accuracy or faster prediction, so that the driverMThe driver should be able to pass by monitoring the driver side, which is easy to graspMTo reduce the burden. The own vehicle A itself also has a narrow road guide system 30 and a driver on the left and right.MIt is possible to safely pass through narrow paths. When the narrow road is taken off, the normal road is judged and the normal system is restored.
[0040]
Although the embodiment of the present invention has been described above, it is a matter of course that the possibility of contact with an obstacle on the right side of the passenger seat is predicted when the left steering wheel can pass through a narrow road. The determination method of a narrow road and the prediction of the possibility of contact when the vehicle can pass through a narrow road are not limited to the embodiment. The image recognized on a part of the screen can be applied other than the possibility of contact on the left side of the vehicle.
[0041]
【The invention's effect】
  As explained above,According to the present inventionIn the narrow road guide device, an image recognition unit that obtains a scene captured in front of the vehicle as three-dimensional image data of a distance distribution from an image signal captured by a front monitoring camera, and a road shape ahead by the image data of the three-dimensional image The road, obstacle detection means for detecting obstacles on the road, and the relationship between the actual road width of the road and the width of the vehicle based on the road shape and obstacle data. A narrow road judging means for judging whether or not, a traffic judging means for judging whether or not it is possible to pass when the narrow road is judged, and a first warning instruction is issued when it is judged that the narrow road is impossible to pass, When it is determined that the road is passable, a second warning command is issued to monitor the vehicle so that it does not come in contact with an obstacle, and the obstacle on the passenger seat side of the vehicle is detected based on the three-dimensional image data of the scenery taken in front. Recognize only the objects and A structure and a vehicle guide means for monitoring only potential catalyst. As a result, narrow roads with narrow roads can be accurately determined in advance, so if the narrow roads are inaccessible, an alarm will be given prior to entry, and if they can pass, only the obstacles on the passenger side of the vehicle will be contacted. Since the possibility is concentrated and predicted and monitored, the calculation processing is reduced by half because the image data to be processed is reduced, and the density of the image data to be processed is doubled. Therefore, the possibility of contact with the obstacle on the passenger seat side can be predicted more quickly and accurately.
[0042]
  Also,By comparing the current position of the obstacle on the passenger seat side and the position after a few seconds, the possibility of contact between the obstacle on the passenger seat side and the passenger seat side of the host vehicle is predicted. As a result, the passenger's side, which is difficult to grasp the own vehicle, is guided so as to safely pass through the narrow road while being monitored so as not to come in contact with the obstacle, so that the burden on the driver can be reduced.
[0043]
  Also,When it is determined that the narrow road can be passed and the driver's deceleration operation is determined, three-dimensional image data of only a partial area in front of the vehicle is obtained. As a result, if the driver is in a situation where he / she is confused about the passage of the narrow road, the passenger seat side is monitored with high accuracy, so that the passage of the narrow road can be performed more appropriately.
[0044]
  Also,The narrow road determination means determines whether or not the road is a narrow road that can pass through the relationship between the road width in front of the host vehicle detected by the detection means and the width of the host vehicle. Further, when it is determined that the narrow road can be passed by the narrow road determination means, only the possibility of contact with the obstacle on the passenger seat side of the own vehicle is monitored by the vehicle guidance means. Thereby, since the accuracy of monitoring on the passenger seat side is improved, the possibility of contact with the obstacle on the passenger seat side can be predicted more quickly and accurately.
[0045]
  Also,When the driver's deceleration operation is determined, the vehicle guide means monitors the possibility of contact with an obstacle on the passenger seat side of the host vehicle. As a result, when the driver is in a situation where he / she is confused with narrow roads, the passenger seat side is monitored with high accuracy, so that the narrow roads can be properly passed.
  Also,A narrow road is determined by the narrow road determination means based on the road shape, the substantial road width of the road based on the obstacle data, and the width of the host vehicle. This makes it possible to accurately and accurately determine whether or not the narrow road is passable, and further improve the prevention of entry into the impassable narrow road and contact with obstacles due to the entry.
[Brief description of the drawings]
FIG. 1 is a view corresponding to a claim showing a configuration of a narrow path guide device according to the present invention.
FIG. 2 is an explanatory diagram showing an outline of an ADA system.
FIG. 3 is a functional block diagram of an embodiment of the present invention.
FIG. 4 is a flowchart of narrow road alarm control.
FIG. 5 is an explanatory view showing an operating state of narrow road traffic.
[Explanation of symbols]
11 CCD camera (front monitoring camera)
13,14 Alarm
21 Image recognition means
22 Road, obstacle detection means
31 Narrow path judging means
32 Traffic determination means
33 Vehicle guidance means

Claims (4)

前方監視カメラで撮像した画像信号で、車両前方の撮像された風景を距離分布の三次元画像データとして得る画像認識手段と、
この三次元画像の画像データにより前方の道路形状を認識すると共に、道路上の障害物を検出する道路、障害物検出手段と、
道路形状と障害物のデータによる道路の実質的な道幅と自車両の幅との関係により、事前に狭路か否かを判定する狭路判定手段と、
前記狭路判定手段により狭路判定されたときに通行可能か否かを判定する通行判定手段と、
前記通行判定手段により狭路の通行不可が判定されると第1の警報指令を発し、前記通行判定手段により狭路の通行可能が判定されると第2の警報指令を発すると共に、前記通行判定手段により狭路の通行可能が判定され、且つドライバの減速操作が判断された場合に、自車両前方における助手席側の一部域のみの三次元画像データを得るように前記画像認識手段を切換えて、前方の撮像された風景の三次元画像データにより自車両の助手席側の障害物のみを認識し、この障害物との接触の可能性のみを監視する車両案内手段とを備えることを特徴とする狭路ガイド装置。
An image recognition means for obtaining, as an image signal captured by a front monitoring camera, a three-dimensional image data of a distance distribution of a scene captured in front of the vehicle;
While recognizing the shape of the road ahead by the image data of this three-dimensional image, the road for detecting obstacles on the road, obstacle detection means,
Narrow road determination means for determining in advance whether or not the road is narrow according to the relationship between the substantial road width of the road and the width of the host vehicle based on the road shape and obstacle data;
Passage determination means for determining whether or not passage is possible when the narrow road is determined by the narrow road determination means;
The traffic determination unit emits the narrow road impassable for is determined first alarm command by, along with the narrow road of passable is determined emit second warning command by the passing determining means, the passage determination The image recognition means is switched so as to obtain three-dimensional image data of only a partial area on the passenger seat side in front of the host vehicle when it is determined that the narrow road can be passed by the means and the driver's deceleration operation is determined. Te, that recognizes only obstacle the passenger side of the vehicle by the three-dimensional image data of the scene that is in front of the imaging, obtain Preparations and vehicle guidance means for monitoring only possibility of contact between the obstacle A narrow path guide device.
前記車両案内手段は、前記助手席側の障害物の現時点と数秒後との位置を比較し、前記自車両の助手席側と助手席側の障害物との接触の可能性を予測することを特徴とする請求項1記載の狭路ガイド装置。  The vehicle guide means compares the current position of the obstacle on the passenger seat side with a few seconds later, and predicts the possibility of contact between the obstacle on the passenger seat side and the passenger seat side of the host vehicle. The narrow path guide device according to claim 1, wherein 車両前方の狭路、その狭路の通行や障害物との接触の可能性を事前に判定して車両を安全にガイドする狭路ガイド装置において、
自車両前方の道路幅を検出する検出手段と、
前記検出手段が検出した道路幅と自車両の幅との関係により通行が可能な狭路か否かを判定する狭路判定手段と、
前記狭路判定手段が通行可能な狭路と判定し、ドライバの減速操作が判断された場合に、前方のうち自車両の助手席側の障害物との接触の可能性のみを監視する車両案内手段とを備えることを特徴とする狭路ガイド装置。
In a narrow road guide device that guides the vehicle safely by determining in advance the narrow road in front of the vehicle, the possibility of passage through the narrow road and contact with obstacles,
Detecting means for detecting a road width in front of the host vehicle;
Narrow road determination means for determining whether or not the road can be passed by the relationship between the road width detected by the detection means and the width of the host vehicle;
Vehicle guidance for monitoring only the possibility of contact with an obstacle on the passenger's seat side of the front of the vehicle when it is determined that the narrow road determination means is a narrow road and the driver's deceleration operation is determined narrow road guide apparatus characterized by obtaining Bei and means.
前記狭路判定手段は、道路形状、障害物のデータによる道路の実質的な道幅及び自車両の幅に基づいて狭路を判定することを特徴とする請求項3記載の狭路ガイド装置。  4. The narrow road guide device according to claim 3, wherein the narrow road determination means determines a narrow road based on a road shape, a substantial road width of the road based on obstacle data, and a width of the host vehicle.
JP28713995A 1995-11-06 1995-11-06 Narrow path guide device Expired - Fee Related JP3881048B2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP28713995A JP3881048B2 (en) 1995-11-06 1995-11-06 Narrow path guide device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP28713995A JP3881048B2 (en) 1995-11-06 1995-11-06 Narrow path guide device

Publications (2)

Publication Number Publication Date
JPH09128687A JPH09128687A (en) 1997-05-16
JP3881048B2 true JP3881048B2 (en) 2007-02-14

Family

ID=17713580

Family Applications (1)

Application Number Title Priority Date Filing Date
JP28713995A Expired - Fee Related JP3881048B2 (en) 1995-11-06 1995-11-06 Narrow path guide device

Country Status (1)

Country Link
JP (1) JP3881048B2 (en)

Families Citing this family (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
FR2776596B1 (en) * 1998-03-24 2000-05-05 Renault Vehicules Ind CAMERA ANTEVISION SYSTEM FOR INDUSTRIAL VEHICLE
JP4657495B2 (en) * 2001-05-29 2011-03-23 富士重工業株式会社 Vehicle driving support device
JP4813304B2 (en) * 2006-09-19 2011-11-09 本田技研工業株式会社 Vehicle periphery monitoring device
JP4730406B2 (en) * 2008-07-11 2011-07-20 トヨタ自動車株式会社 Driving support control device
JP5516301B2 (en) 2010-10-05 2014-06-11 トヨタ自動車株式会社 Vehicle runway determination system
JP5796947B2 (en) * 2010-10-08 2015-10-21 三菱重工業株式会社 Autonomous traveling control device and autonomous traveling vehicle equipped with the same
JP5408240B2 (en) * 2011-12-12 2014-02-05 株式会社デンソー Warning system, vehicle device, and server
JP6001901B2 (en) * 2012-03-29 2016-10-05 富士重工業株式会社 Hybrid vehicle travel control device
KR20130130557A (en) * 2012-05-22 2013-12-02 현대모비스 주식회사 Apparatus and method for assisting wynd pass of vehicle
CN102717816A (en) * 2012-06-21 2012-10-10 中国神华能源股份有限公司 Situation processing system of locomotive operation information
JP5890299B2 (en) * 2012-12-07 2016-03-22 本田技研工業株式会社 Driving assistance device
JP6015505B2 (en) * 2013-03-15 2016-10-26 株式会社デンソー Information display device
CN104670244A (en) * 2015-01-29 2015-06-03 柳州市二和汽车零部件有限公司 Vehicle safety control system with voice control function
CN105480231A (en) * 2015-12-18 2016-04-13 苏州市享乐惠信息科技有限公司 Vehicle self-service driving system
WO2017217420A1 (en) 2016-06-14 2017-12-21 大日本印刷株式会社 Illuminating device, hologram element, and vehicle control method
CN113874268A (en) * 2019-05-22 2021-12-31 日立安斯泰莫株式会社 Vehicle control device
CN111993415B (en) * 2020-07-30 2022-04-19 广州大学 Autonomous patrol robot, mobile monitoring method and device thereof, and autonomous patrol system
CN115171430B (en) * 2022-08-15 2024-03-12 中通服建设有限公司 Intelligent public transport emergency processing system based on image recognition

Also Published As

Publication number Publication date
JPH09128687A (en) 1997-05-16

Similar Documents

Publication Publication Date Title
JP3881048B2 (en) Narrow path guide device
EP0896918B1 (en) Drive assist system for motor vehicles
JP6115576B2 (en) Vehicle travel control device
JP6180968B2 (en) Vehicle control device
JP4525915B2 (en) Driving assistance device
JP4309843B2 (en) Method and apparatus for preventing vehicle collision
US7561180B2 (en) Movable body safety system and movable body operation support method
JP5316706B2 (en) Emergency vehicle evacuation device
JP4886510B2 (en) Method and apparatus for determining vehicle position and / or intended position relative to opposite lanes of a multi-lane roadway during a parking process
US10173680B2 (en) Vehicle speed control device
KR101864896B1 (en) Method of capturing the surroundings of a vehicle
JPH08205306A (en) Alarm device for car
JPH1153694A (en) Intersection warning device
JP6085958B2 (en) Parking assistance device
US11279352B2 (en) Vehicle control device
JP4952938B2 (en) Vehicle travel support device
JPH1069598A (en) Collision preventing device for vehicle
CN110225853A (en) It is avoided collision with cross traffic
JP2004280453A (en) Vehicular right turn safety confirming system
JPH1153695A (en) Curved road warning device
JPH09263200A (en) Alarm device for vehicle
KR20230089780A (en) Method and apparatus for collision avoidance
KR20150130056A (en) Apparatus for making a driver operate a car easily and control method thereof
KR20220071560A (en) Apparatus amd method for warning pedestrian in emergency breaking situation
JPH09254726A (en) Operation assisting device for vehicle

Legal Events

Date Code Title Description
A977 Report on retrieval

Free format text: JAPANESE INTERMEDIATE CODE: A971007

Effective date: 20040824

A131 Notification of reasons for refusal

Free format text: JAPANESE INTERMEDIATE CODE: A131

Effective date: 20041207

A521 Written amendment

Free format text: JAPANESE INTERMEDIATE CODE: A523

Effective date: 20050128

A131 Notification of reasons for refusal

Free format text: JAPANESE INTERMEDIATE CODE: A131

Effective date: 20060210

A521 Written amendment

Free format text: JAPANESE INTERMEDIATE CODE: A523

Effective date: 20060410

A131 Notification of reasons for refusal

Free format text: JAPANESE INTERMEDIATE CODE: A131

Effective date: 20060825

A521 Written amendment

Free format text: JAPANESE INTERMEDIATE CODE: A523

Effective date: 20060829

TRDD Decision of grant or rejection written
A01 Written decision to grant a patent or to grant a registration (utility model)

Free format text: JAPANESE INTERMEDIATE CODE: A01

Effective date: 20061107

A61 First payment of annual fees (during grant procedure)

Free format text: JAPANESE INTERMEDIATE CODE: A61

Effective date: 20061109

R150 Certificate of patent or registration of utility model

Free format text: JAPANESE INTERMEDIATE CODE: R150

FPAY Renewal fee payment (event date is renewal date of database)

Free format text: PAYMENT UNTIL: 20091117

Year of fee payment: 3

FPAY Renewal fee payment (event date is renewal date of database)

Free format text: PAYMENT UNTIL: 20101117

Year of fee payment: 4

FPAY Renewal fee payment (event date is renewal date of database)

Free format text: PAYMENT UNTIL: 20111117

Year of fee payment: 5

FPAY Renewal fee payment (event date is renewal date of database)

Free format text: PAYMENT UNTIL: 20121117

Year of fee payment: 6

FPAY Renewal fee payment (event date is renewal date of database)

Free format text: PAYMENT UNTIL: 20121117

Year of fee payment: 6

FPAY Renewal fee payment (event date is renewal date of database)

Free format text: PAYMENT UNTIL: 20131117

Year of fee payment: 7

R250 Receipt of annual fees

Free format text: JAPANESE INTERMEDIATE CODE: R250

S531 Written request for registration of change of domicile

Free format text: JAPANESE INTERMEDIATE CODE: R313531

R350 Written notification of registration of transfer

Free format text: JAPANESE INTERMEDIATE CODE: R350

LAPS Cancellation because of no payment of annual fees