JP3743071B2 - Lane departure warning device - Google Patents

Description

【００２１】
道路マーカ地図メモリ１５は、高速道路の路線、区間ごとに対応して、地点の番号、埋め込まれた磁気ネイルのその地点の極の区別（「磁気ネイルの符号」という）、東向きをｘ方向とし北向きをｙ方向とした相対座標を記憶している。例えば、出発点となる地点の番号は０、その地点に埋め込まれた磁気ネイルの符号はＳ、地点の相対座標は（０，０）となっている。
【００２２】
位置推定部１４は、走行距離検出部１１から出力される距離データΔＤ_n を取り込むとともに、走行方向検出部１２から出力される角速度データω_n を取り込み、前回の角速度ω_n-1 と今回の角速度ω_n との平均Ω_n
Ω_n ＝（ω_n-1 ＋ω_n ）／２
を求め、これに周期Δｔをかけ、前回の方位θ_n-1 に加算することにより今回の方位θ_n を求める。
【００２３】
θ_n ＝Ω_n Δｔ＋θ_n-1
さらに、前回の方位θ_n-1 と、今回の方位θ_n との平均方位Θ_n を求め、
Θ_n ＝（θ_n ＋θ_n-1 ）／２
そしてこの平均方位Θ_n に基づいて、前記距離ΔＤ_n の東西方向成分Δｘ_n （＝ΔＤ_n × cosΘ_n ）、および南北方向成分Δｙ_n （＝ΔＤ_n × sinΘ_n ）を検出し、前回の推定位置データ（ｘ_n-1 ，ｙ_n-1 ）に対して前記各成分Δｘ_n ，Δｙ_n を加算することにより、現在の推定位置（ｘ_n ，ｙ_n ）を検出する。
【００２４】
ｘ_n ＝ｘ_n-1 ＋Δｘ_n
ｙ_n ＝ｙ_n-1 ＋Δｙ_n
なお、推定位置の初期位置（ｘ₀ ，ｙ₀ ）を決める必要があるが、この初期位置（ｘ₀ ，ｙ₀ ）は、出発点となる磁気ネイルを検出した時点の車両位置を、例えば（０，０）に設定してそれを採用すればよい。
【００２５】
位置補正部１６は、磁気抵抗素子の取付け長さを超えて車線を逸脱していないことを前提として、この推定位置（ｘ_n ，ｙ_n ）を、道路マーカ地図メモリ１５に格納された相対座標に基づいて補正する。
この補正方法を説明すると、まず、コース出発点を特定する。
ここで、コース出発点の特定は、次のようにして行う。車両は、高速道路に入る前に通信手段、例えばビーコンから、当該高速道路の路線名、区間名の情報を受ける。ナビゲーション装置を搭載していれば、ナビゲーション装置から路線名、区間名の情報を受けてもよい。この情報により、道路マーカ地図メモリ１５に記憶されているデータ群の記憶場所を特定することができる。
【００２６】
高速道路の入口と出口との間に磁気ネイルが埋め込まれており、車両が当該入口から高速道路に入ると、最初は特定の符号（例えばＳ極）が現れ、その後両符号が所定のコードに従って現れる。この最初の特定の符号を初めて検出した地点を出発点とする。その後、磁気ネイルにより構成されるコードを読み取ることにより、相対座標を知ることができる。
【００２７】
また、例えばコードが−１，１，−１，１，‥‥といった単純なものであれば、検出した符号をカウントするだけで、何番目の地点であるかが分かるので、表１を参照して相対座標を知ることができる。
位置補正部１６は前記コードの読み取り結果に基づき、道路マーカ地図メモリ１５を参照して相対座標（以下「（ｘ_m ，ｙ_m ）」と書く）を読み出す。
【００２８】
次に、当該時点の車両推定位置（ｘ_s ，ｙ_s ）を求める。一般に、当該車両推定位置（ｘ_s ，ｙ_s ）は、図４に示すようにいずれかの検出周期Δｔの始めの車両推定位置（ｘ_n-1 ，ｙ_n-1 ）と終わりの車両推定位置（ｘ_n ，ｙ_n ）との間に存在するので、車両推定位置（ｘ_s ，ｙ_s ）は、検出周期Δｔの始めの時刻Ｔ_n-1 と、終わりの時刻Ｔ_n と、符号が検出された時刻Ｔ_m とを使えば、
ｘ_s ＝ｘ_n-1 ＋（ｘ_n −ｘ_n-1 ）（Ｔ_m −Ｔ_n-1 ）／Δｔ
ｙ_s ＝ｙ_n-1 ＋（ｙ_n −ｙ_n-1 ）（Ｔ_m −Ｔ_n-1 ）／Δｔ
で求められる。
【００２９】
そして、当該車両推定位置（ｘ_s ，ｙ_s ）及び前記相対距離Ｌ_m に基づいて推定される相対座標（ｘ′_m ，ｙ′_m ）と、道路マーカ地図メモリ１５に記憶されている相対座標（ｘ_m ，ｙ_m ）との関係を求める。図５は、車両推定位置（ｘ_s ，ｙ_s ）と推定位置（ｘ′_m ，ｙ′_m ）との位置関係を表した図であり、車両の走行方向はθ_n で表されている。方向θ_n の符号は反時計回りを正とし、相対距離Ｌ_m の符号は車両の左側に磁気テープ又は誘導線を検出したときを正とする。同図に示した関係より、推定位置（ｘ′_m ，ｙ′_m ）は、Ｌ_m とθ_n とを用いて
ｘ′_m ＝ｘ_s −Ｌ_m sin θ_n (1)
ｙ′_m ＝ｙ_s ＋Ｌ_m cos θ_n (2)
で表される。
【００３０】
ここで、前記(1) 式及び(2) 式で示された推定位置（ｘ′_m ，ｙ′_m ）と、道路マーカ地図メモリ１５から読みだした磁気ネイルの座標（ｘ_m ，ｙ_m ）とを比較し、その差を求める。
Δｘ＝ｘ_m −ｘ_s
Δｙ＝ｙ_m −ｙ_s
この差Δｘ，Δｙが車両推定位置（ｘ_n ，ｙ_n ）の補正量となる（以下、添字ｎを省略して、推定位置（ｘ，ｙ）と書く）。すなわち、車両推定位置（ｘ，ｙ）を、
ｘ＝ｘ＋Δｘ
ｙ＝ｙ＋Δｙ
として補正することができる。補正された推定位置（ｘ，ｙ）は、ＧＰＳ(Global Positioning System) 等と比べて極めて精度が高い。
【００３１】
前述した位置補正部１６の処理は、磁気抵抗素子の取付け長さを超えて車線を逸脱していないことを前提としていた。
もし、磁気抵抗素子の取付け長さを超えて車線を逸脱すれば、車両の中心線と磁気テープとの横方向の相対距離Ｌ_m が求まらないので、前述した処理を行うことはできない。この場合は、位置推定部１４から出力される車両の推定位置（ｘ_n ，ｙ_n ）をそのまま車両の推定位置（ｘ，ｙ）として採用することになる。
【００３２】
次に、車線からの逸脱を警告する方法について説明する。
この警告の判断は、車両が、磁気抵抗素子の取付け長さを超えて車線を逸脱しても、逸脱していなくても行うことができる（逸脱していないのに警告するのは無用という気もするが、警告しきい値は任意に設定することができる）。
まず、逸脱検出部１９は、車両推定位置（ｘ，ｙ）を磁気ネイルの並んでいるライン上の位置へ投影する。
【００３３】
図６は、最近検出された磁気ネイルの座標（ｘ_m0，ｙ_m0）、次の番号に対応する磁気ネイルの座標（ｘ_m1，ｙ_m1）及び推定位置（ｘ，ｙ）を示す図である。
ライン上の補正すべき位置を（ｘ_b ，ｙ_b ）と書くことにすると、（ｘ_b ，ｙ_b ）は、推定位置（ｘ，ｙ）からラインに垂線を引くことにより求められる。
ｘ_b ＝ｘ_m0＋（ｘ_m1−ｘ_m0）Ｒ
ｙ_b ＝ｙ_m0＋（ｙ_m1−ｙ_m0）Ｒ
ここで、Ｒは、
Ｒ＝｛（ｘ_m1−ｘ_m0）（ｘ−ｘ_m0）＋（ｙ_m1−ｙ_m0）（ｙ−ｙ_m0）｝
／｛（ｘ_m1−ｘ_m0）² ＋（ｙ_m1−ｙ_m0）² ｝
で定義される。
【００３４】
このライン上への投影位置（ｘ_b ，ｙ_b ）と、車両推定位置（ｘ，ｙ）との距離ｄ
ｄ＝ＳＱＲＴ｛（ｘ−ｘ_b ）² ＋（ｙ−ｙ_b ）² ｝
が車両のラインからのずれになる。
逸脱検出部１９は、このずれｄが閾値ｄ₀ よりも大きいと判断すれば、逸脱警告部２０から音声により警告を出す。勿論、音声で警告を出すのに代えて、ディスプレイ（図示せず）に警告を表示させてもよい。また、警告を出すのと同時に、ブレーキ操作、アクセル操作、ステアリング操作等を行うアクチュエータに指令信号を与えてもよい。
【００３５】
警告の解除は、ウィンカー操作があったとき、ステアリング操作があったときとする。また、ウィンカー操作、ステアリング操作に伴ってラインからのずれｄが閾値ｄ₀ を越えたときは、初めから警告をしないようにしてもよい。
なお、前記閾値ｄ₀ は、例えば、車両の幅、車線幅を考慮して選ぶことができる。この場合、車両の幅は、当該車両に固有の値として保有しておけばよく、車線幅は、高速道路の路線、区間ごとに対応して、データとして道路マーカ地図メモリ１５に格納しておけばよい。
【００３６】
例えば、車線幅３．５ｍ、車両の幅１．８ｍのときは、閾値ｄ₀ を０．８５ｍに設定する。
閾値ｄ₀ の設定をするとき、図７に示すように、車両の走行方向と磁気ネイルの並んでいるラインとのヨー角φを考慮してもよい。ここで、ヨー角φは、道路マーカからのずれｄの時間変化に基づいて算出することができる。ヨー角φが大きいときは、ラインから離れるスピードが速いことを意味するので、閾値を小さく設定する。ヨー角φが小さいときは、ハンドルの切り返しも容易なので警告が遅くなってもよいと考え閾値を大きめに設定する。
【００３７】
具体的にいうと、一般に車両が直線維持走行を行っているときは、ヨー角φは±１．５°以内、車線変更を行うときはヨー角φは３°〜５°程度になる。したがって、ヨー角φが１．５°未満のときは車線の境界を示す白線ぎりぎりで警告を与え、ヨー角φが１．５°以上３°未満のときは白線の３０ｃｍ内側で警告を与え、ヨー角φが３°以上のときは白線の５０ｃｍ内側で警告を与えるようにする。
【００３８】
また、閾値ｄ₀ を車両の幅、車線幅だけでなく、ガードレール又は側壁までの距離を考慮して決定することもできる。この場合、白線とガードレール又は側壁までの距離を道路マーカ地図メモリ１５に格納しておき、この距離が比較的短いときは閾値ｄ₀ を小さめにし、この距離が比較的長いときは閾値ｄ₀ を大きめにする。
【００３９】
さらに、ドライバの運転の癖によって、閾値を変えることもできる。ドライバによっては、左寄りを走行する人、右寄りを走行する人がいる。この場合は、過去の走行内容を学習しておき、右側と左側とで閾値を変える。
また、車両の性能等により、左右のぶれが大き目の場合があるので、この場合も、過去の走行内容を学習しておき、左右のぶれが大きいときは閾値を小さめに設定する。
【００４０】
【発明の効果】
以上のように本発明の車線逸脱警告装置によれば、自律航法により求められた車両の位置を道路マーカの並ぶライン上に投影することにより、車両のラインからのずれｄを求めることができる。このずれｄに基づいて車線の逸脱を的確に判断し警告することができる。
【００４１】
特に、請求項２記載の発明であれば、自律航法により求められた車両の位置を、道路マーカ地図メモリに記憶された道路マーカの座標データ列に基づいて、正確に補正することができ、車両のラインからのずれｄを正確に求めることができる。
【図面の簡単な説明】
【図１】道路マーカとして、車線の真中に磁気ネイルを埋め込んだ道路を低空から見た図である。
【図２】車線逸脱警告装置を含む車載装置を示すブロック図である。
【図３】磁気抵抗素子を細長いゴム状の永久磁石の上に複数個並べ、それぞれから出力電圧を取り出す回路構成を示す図である。
【図４】いずれかの検出周期Δｔの始めの車両推定位置（ｘ_n-1 ，ｙ_n-1 ）と終わりの車両推定位置（ｘ_n ，ｙ_n ）との間の車両推定位置（ｘ_s ，ｙ_s ）の算出方法を説明する図である。
【図５】車両推定位置（ｘ_s ，ｙ_s ）と磁気ネイルの座標（ｘ_m ，ｙ_m ）との位置関係を説明する図である。
【図６】車両推定位置（ｘ，ｙ）を磁気ネイルの並んでいるライン上の位置へ投影する方法を説明する図である。
【図７】車両推定位置（ｘ，ｙ）を磁気ネイルの並んでいるライン上の位置へ投影する場合にヨー角φを考慮して投影する方法を説明する図である。
【符号の説明】
１ 磁気抵抗素子
２ 抵抗素子
３ 永久磁石
１１ 走行距離検出部
１２ 走行方向検出部
１３ 磁界検出部
１３ａ 波形整形部
１４ 位置推定部
１５ 道路マーカ地図メモリ
１６ 位置補正部
１７ アンテナ
１７ａ 受信復調部
１７ｂ ループコイル
１７ｃ ループコイル
１７ｄ ９０°移相器
１７ｅ ハイブリッド[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a lane departure warning device that lays out a road marker on a road and detects a deviation of the vehicle from the road marker by detecting the road marker to warn of departure from the lane.
[0002]
[Prior art]
By using the communication between the road and the vehicle or the communication between the vehicle and the vehicle, the lane position of the vehicle, the travel distance, the speed, the acceleration, the inter-vehicle distance from other vehicles, etc. are taken as information, and the driver is warned based on this information, 2. Description of the Related Art There is known an automatic driving system that generates commands such as a vehicle brake operation, an accelerator operation, and a steering operation and transmits them to a driving device.
[0003]
In such an automatic driving system, the travel distance of the vehicle (relative travel distance from a predetermined reference point), which lane of the road the vehicle is traveling on, whether it is not off the lane, or which lane It is important to accurately recognize whether the vehicle is traveling in the part (center of lane, left side or right side), running straight along the lane, or crossing diagonally.
[0004]
Conventionally, a technique for determining a traveling position by photographing a road state through an in-vehicle camera and recognizing an image has been proposed (see, for example, JP-A-7-77431), but a camera and an image processing device are necessary. Therefore, since the determination algorithm is complicated, an apparatus that can more easily recognize the traveling position has been desired.
Therefore, a system has been proposed in which a magnetic nail is embedded in the road surface along the direction of the road, and this magnetic nail is detected by a magnetic sensor provided on the vehicle body, thereby detecting the travel distance of the vehicle and the lateral displacement of the vehicle body. (See US Pat. No. 5,347,456 on September 13, 1994, PCT International Publication WO 92/08176 on May 14, 1992).
[0005]
Further, a system has been proposed in which a magnetic tape is attached to a road and the position of the vehicle is detected using a magnetic field generated from the magnetic tape.
The equipment provided on the road in order to detect the traveling position of the vehicle, such as these magnetic nails and magnetic tape, will be collectively referred to as “road marker”.
A vehicle equipped with an automatic driving system can know the traveling position of the vehicle by detecting the road marker as described above.
[0006]
[Problems to be solved by the invention]
By the way, in order to accurately recognize whether it is not off the lane, or which part of the lane is running, whether it is running straight along the lane or crossing diagonally, etc. It is necessary to recognize the deviation.
The most common method is to arrange a plurality of detection elements for detecting a road marker on the bottom of a vehicle bumper, etc., and calculate the deviation from the road marker depending on which detection element detects the road marker. Is the method.
[0007]
For example, if the detection elements are attached to the left and right and middle of the bottom of the bumper, and the detection element detects a road marker in the middle, it is assumed that the vehicle is traveling in the center of the lane, and the left detection element detects the road marker. It is assumed that the vehicle is traveling on the right side of the lane when the vehicle is on the road, and the vehicle is traveling on the left side of the lane when the right detection element detects the road marker.
However, in the detection method described above, when the vehicle deviates from the lane beyond the attachment length of the detection element, the deviation from the road marker is not known. Moreover, the mounting length of the detection element is limited to the width of the vehicle body.
[0008]
On the other hand, autonomous navigation is known in which an estimated position of a vehicle is calculated based on a vehicle travel distance detection signal and a travel direction detection signal.
Accordingly, an object of the present invention is to realize a lane departure warning device capable of accurately determining and warning a lane departure by storing a coordinate data string of road markers and using autonomous navigation together. .
[0009]
[Means for Solving the Problems]
A lane departure warning device according to the present invention includes a vehicle position estimating means for estimating the position of a vehicle by autonomous navigation, a road marker map memory storing a coordinate data string of road markers, and the vehicle position on a line on which road markers are arranged. A deviation d from the line of the vehicle is obtained by projection, and a deviation detecting means for detecting a lane deviation is detected by comparing the deviation d with a threshold value d ₀ and a deviation warning means. 1).
[0010]
According to this configuration, the deviation d from the line of the vehicle can be obtained by projecting the position of the vehicle obtained by the autonomous navigation onto the line where the road markers are arranged. If this deviation is greater than or equal to the threshold d ₀ , a lane departure can be detected and alerted.
The lane departure warning device according to the present invention includes a marker detection means for detecting a road marker, a road marker detected by the marker detection means, and a road stored in a coordinate data memory, in addition to the elements described in claim 1. The estimated position of the road marker detected by the marker detecting means is calculated based on the vehicle position estimated by the vehicle position estimating means in association with the marker, and the estimated position of the road marker and the coordinate data are calculated. The apparatus further comprises position correcting means for correcting the position of the vehicle estimated by autonomous navigation by comparing with the coordinate data of the road marker stored in the memory (claim 2).
[0011]
According to this configuration, the position of the vehicle obtained by autonomous navigation can be accurately corrected based on the coordinate data string of the road marker stored in the road marker map memory. Therefore, the deviation d from the vehicle line can be accurately obtained.
The threshold value d ₀ can be set based on any one of (a) to (d) described in claim 3 or a combination thereof (claim 3).
[0012]
The reasons for adopting the criteria (a) to (d) are as follows.
(a): since easily depart from the lane width of the vehicle is large, reduce the threshold d _0.
(b): Since the lane width is likely to deviate from the narrow and lanes to reduce the threshold d _0.
(c): When the yaw angle φ is large, it means that the speed of moving away from the line is fast. Therefore, it is dangerous for driving unless warning is given early, so the threshold is set small. When the yaw angle φ is small, it is easy to turn the steering wheel, so the warning may be delayed, and the threshold value is set larger.
(d): the distance from the white line defining a lane to guardrail or sidewall is short, there is a risk of collision and not issue a warning to the early, to smaller threshold d _0. When this distance is relatively long to large threshold d _0.
[0013]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.
FIG. 1 is a view of a road in which a magnetic nail is embedded in the middle of a lane as a road marker, viewed from a low sky.
The magnetic nail is composed of a ferromagnetic nail having a diameter of about 3 cm and a length of about 10 cm.
[0014]
A magnetic nail has N poles and S poles with a predetermined pitch L (for example, 1 to 2 m) as a period, and a predetermined code (for example, 1 is N pole and -1 is S pole, -1, 1, -1, 1,... The polarity means the polarity of the magnetic field facing the ground surface.
FIG. 2 is a block diagram showing an in-vehicle device including a lane departure warning device. The vehicle-mounted device calculates the estimated position by autonomous navigation based on the travel distance detection unit 11, the travel direction detection unit 12, the travel distance detection signal output from the both detection units 11 and 12, and the travel direction detection signal. Unit 14, a magnetic field detection unit 13 and a waveform shaping unit 13a for detecting the magnetic field of the magnetic nail embedded in the road, a road marker map memory 15 storing a code and a coordinate data string formed by the magnetic nail, A position correction unit 16 that detects a magnetic nail code and compares the code with the coordinate data stored in the road marker map memory 15 to correct the estimated position detected by the position estimation unit 14; A section 19 and a deviation warning section 20 are provided.
[0015]
The position estimation unit 14, the road marker map memory 15, and the position correction unit 16 constitute a position detection unit 18.
More specifically, for example, a wheel speed sensor can be used as the travel distance detector 11. This wheel speed sensor has a magnetic sensor that detects the rotation of the wheel, and counts the wave number of the output sine wave signal from the magnetic sensor by the counter to obtain the wheel rotation speed, and outputs the count data output from the counter. On the other hand, the travel distance per detection period Δt (period number is represented by n) is detected by multiplying a predetermined constant indicating the outer periphery of the wheel by a multiplier. In addition, a conventionally known configuration such as a vehicle speed sensor configured to detect a travel distance by detecting a travel speed of a vehicle based on a Doppler shift or the like and integrating it can be used.
[0016]
For the traveling direction detection unit 12, for example, a gyro that outputs rotation angle data per unit time can be used. Examples of the gyro include a vibration gyro, an optical fiber gyro, and a differential wheel speed sensor. Further, a geomagnetic sensor that detects the horizontal component of geomagnetism can be used, and a combination of a geomagnetic sensor and the gyro can also be adopted.
[0017]
The magnetic field detector 13 is provided, for example, at the bottom of a bumper. As shown in FIG. 3, a plurality of magnetoresistive elements 1 are arranged on an elongated rubber-like permanent magnet 3, and an output voltage is extracted from each of them. Yes.
The magnetoresistive element 1 is an element whose electric resistance changes when a magnetic field is applied, and a semiconductor material such as InSb, GaAs, InAs or the like is often used. In the present embodiment, an element in which an InSb thin film is formed on a substrate by a vacuum deposition method is used (for example, “Magnetoresistive element TMS-D series” manufactured by Toyocom Toyo Tsushinki Co., Ltd. can be used). In addition, a bulk (single crystal) type magnetoresistive element may be used.
[0018]
If any one of the magnetoresistive elements of the magnetic field detector 13 detects the magnetic field of the magnetic nail embedded along the road, the code of the magnetic nail can be read by the change. Further, by detecting which magnetoresistive element of the magnetic field detector 13 is detecting, the relative distance in the lateral direction between the center line of the vehicle and the magnetic tape (hereinafter referred to as “L _m ”) is detected. You can also.
[0019]
The road marker map memory 15 stores codes and position information composed of magnetic nails.
Here, the storage structure of the road marker map memory 15 is as shown in Table 1.
[0020]
[Table 1]

[0021]
The road marker map memory 15 corresponds to each expressway route and section, the point number, the discrimination of the pole of the embedded magnetic nail (referred to as “magnetic nail code”), and the east direction in the x direction. Relative coordinates with the north direction as the y direction are stored. For example, the number of the starting point is 0, the sign of the magnetic nail embedded at that point is S, and the relative coordinates of the point are (0, 0).
[0022]
The position estimator 14 captures the distance data ΔD _n output from the travel distance detector 11 and the angular velocity data ω _n output from the travel direction detector 12 to obtain the previous angular velocity ω _n-1 and the current angular velocity. the average of the ω _n Ω _n
Ω _n = (ω _n-1 + ω _n ) / 2
Is multiplied by the period Δt and added to the previous azimuth θ _n−1 to obtain the current azimuth θ _n .
[0023]
θ _n = Ω _n Δt + θ _n-1
Further, an average azimuth Θ _n between the previous azimuth θ _n-1 and the current azimuth θ _n is obtained,
Θ _n = (θ _n + θ _n-1 ) / 2
Then, based on the average orientation theta _n, the distance [Delta] D _n east-west direction component [Delta] x _n a _{(= ΔD n × cosΘ n)} , and detects a north-south direction component _{Δy n (= ΔD n × sinΘ} n), the previous estimate The current estimated position (x _n , y _n ) is detected by adding the components Δx _n and Δy _n to the position data (x _n−1 , y _n−1 ).
[0024]
x _n = x _n-1 + Δx _n
y _n = y _n-1 + Δy _n
It is necessary to determine the initial position (x ₀ , y ₀ ) of the estimated position. This initial position (x ₀ , y ₀ ) is the vehicle position at the time of detecting the magnetic nail as the starting point, for example ( It may be set to 0,0) and adopted.
[0025]
The position correction unit 16 uses the estimated coordinates (x _n , y _n ) as relative coordinates stored in the road marker map memory 15 on the assumption that it does not deviate from the lane beyond the attachment length of the magnetoresistive element. Correct based on
The correction method will be described. First, the course starting point is specified.
Here, the course starting point is specified as follows. Before entering the expressway, the vehicle receives information on the route name and section name of the expressway from a communication means such as a beacon. If the navigation device is installed, the route name and section name information may be received from the navigation device. With this information, the storage location of the data group stored in the road marker map memory 15 can be specified.
[0026]
Magnetic nails are embedded between the entrance and exit of the highway. When a vehicle enters the highway from the entrance, a specific code (for example, S pole) appears first, and then both codes follow a predetermined code. appear. The point where the first specific code is detected for the first time is taken as the starting point. Thereafter, the relative coordinates can be known by reading a code composed of magnetic nails.
[0027]
For example, if the code is a simple one such as -1, 1, -1, 1,..., It is possible to know the position by simply counting the detected code. To know the relative coordinates.
The position correction unit 16 reads the relative coordinates (hereinafter referred to as “(x _m , y _m )”) with reference to the road marker map memory 15 based on the result of reading the code.
[0028]
Next, the estimated vehicle position (x _s , y _s ) at that time is obtained. In general, the estimated vehicle position (x _s , y _s ) includes the estimated vehicle position (x _n−1 , y _n−1 ) at the beginning and the estimated vehicle position at the end of any detection period Δt as shown in FIG. (X _n , y _n ), the vehicle estimated position (x _s , y _s ) is detected by the sign of the start time T _n-1 and the end time T _n of the detection cycle Δt. If the time _Tm is used,
x _s = x _n-1 + (x _n -x _n-1 ) (T _m -T _n-1 ) / Δt
y _s = y _n−1 + (y _n −y _n−1 ) (T _m −T _n−1 ) / Δt
Is required.
[0029]
Then, relative coordinates (x ′ _m , y ′ _m ) estimated based on the estimated vehicle position (x _s , y _s ) and the relative distance L _m, and relative coordinates stored in the road marker map memory 15. The relationship with (x _m , y _m ) is obtained. FIG. 5 is a diagram showing the positional relationship between the estimated vehicle position (x _s , y _s ) and the estimated position (x ′ _m , y ′ _m ), and the traveling direction of the vehicle is represented by θ _n . The sign of the direction θ _n is positive when it is counterclockwise, and the sign of the relative distance L _m is positive when a magnetic tape or guide wire is detected on the left side of the vehicle. From the relationship shown in the figure, the estimated position (x ′ _m , y ′ _m ) is _expressed as x ′ _m = x _s −L _m sin θ _n (1) using L _m and θ _n.
y ′ _m = y _s + L _m cos θ _n (2)
It is represented by
[0030]
Here, the estimated position (x ′ _m , y ′ _m ) indicated by the above equations (1) and (2) and the coordinates (x _m , y _m ) of the magnetic nail read from the road marker map memory 15 And find the difference.
Δx = x _m −x _s
Δy = y _m −y _s
The difference Δx, Δy becomes the correction amount of the estimated vehicle position (x _n , y _n ) (hereinafter, the subscript n is omitted and the estimated position (x, y) is written). That is, the estimated vehicle position (x, y) is
x = x + Δx
y = y + Δy
As can be corrected. The corrected estimated position (x, y) is extremely accurate as compared with GPS (Global Positioning System) or the like.
[0031]
The processing of the position correction unit 16 described above is based on the assumption that the installation length of the magnetoresistive element has not been exceeded and the vehicle has not deviated from the lane.
If, by deviating from the lane beyond the mounting length of the magnetoresistive element, the lateral relative distance L _m between the center line and the magnetic tape of the vehicle is not determined, it is impossible to perform the processing described above. In this case, the estimated position (x _n , y _n ) of the vehicle output from the position estimation unit 14 is directly adopted as the estimated position (x, y) of the vehicle.
[0032]
Next, a method for warning the departure from the lane will be described.
This warning can be judged whether the vehicle has deviated from the lane beyond the installation length of the magnetoresistive element or not deviated (it is useless to warn that it has not deviated). However, the warning threshold can be set arbitrarily).
First, the deviation detection unit 19 projects the vehicle estimated position (x, y) to a position on a line where magnetic nails are arranged.
[0033]
FIG. 6 is a diagram showing the recently detected magnetic nail coordinates (x _m0 , y _m0 ), the magnetic nail coordinates (x _m1 , y _m1 ) corresponding to the next number, and the estimated position (x, y). .
If the position to be corrected on the line is written as (x _b , y _b ), (x _b , y _b ) can be obtained by drawing a perpendicular to the line from the estimated position (x, y).
x _b = x _m0 + (x _m1 −x _m0 ) R
y _b = y _m0 + (y _m1 −y _m0 ) R
Where R is
R = {(x _m1 −x _m0 ) (x−x _m0 ) + (y _m1 −y _m0 ) (y−y _m0 )}
/ {(X _m1 -x _m0 ) ² + (y _m1 -y _m0 ) ² }
Defined by
[0034]
The distance d between the projection position (x _b , y _b ) on this line and the estimated vehicle position (x, y)
d = SQRT {(x−x _b ) ² + (y−y _b ) ² }
Deviates from the vehicle line.
If the deviation detection unit 19 determines that the deviation d is larger than the threshold value d ₀ , the deviation detection unit 19 issues a warning by voice from the deviation warning unit 20. Of course, instead of issuing a warning by voice, a warning may be displayed on a display (not shown). In addition, a command signal may be given to an actuator that performs a brake operation, an accelerator operation, a steering operation, and the like at the same time as issuing a warning.
[0035]
The warning is released when there is a winker operation or a steering operation. Further, when the deviation d from the line exceeds the threshold value d ₀ due to the blinker operation and the steering operation, the warning may not be given from the beginning.
The threshold value d ₀ can be selected in consideration of the width of the vehicle and the lane width, for example. In this case, the width of the vehicle may be held as a value unique to the vehicle, and the lane width may be stored in the road marker map memory 15 as data corresponding to each route and section of the expressway. That's fine.
[0036]
For example, when the lane width is 3.5 m and the vehicle width is 1.8 m, the threshold value d ₀ is set to 0.85 m.
When setting the threshold value d ₀ , as shown in FIG. 7, the yaw angle φ between the traveling direction of the vehicle and the line along which the magnetic nails are arranged may be taken into consideration. Here, the yaw angle φ can be calculated based on the time change of the deviation d from the road marker. When the yaw angle φ is large, it means that the speed away from the line is fast, so the threshold value is set small. When the yaw angle φ is small, it is easy to turn the steering wheel, so the warning may be delayed, and the threshold value is set larger.
[0037]
Specifically, the yaw angle φ is generally within ± 1.5 ° when the vehicle is traveling in a straight line, and the yaw angle φ is about 3 ° to 5 ° when changing lanes. Therefore, when the yaw angle φ is less than 1.5 °, a warning is given at the margin of the white line indicating the lane boundary, and when the yaw angle φ is 1.5 ° or more and less than 3 °, a warning is given 30 cm inside the white line, When the yaw angle φ is 3 ° or more, a warning is given 50 cm inside the white line.
[0038]
Further, the threshold value d ₀ can be determined in consideration of not only the vehicle width and lane width but also the distance to the guard rail or the side wall. In this case, it may be stored the distance to the white line and the guardrail or sidewalls in the road marker map memory 15, when the distance is relatively short to smaller threshold d _0, the threshold value d ₀ when the distance is relatively long Make it larger.
[0039]
Furthermore, the threshold value can be changed according to the driving habits of the driver. Depending on the driver, there are people who run to the left and people who run to the right. In this case, the past traveling content is learned, and the threshold value is changed between the right side and the left side.
In addition, since the left and right shake may be large depending on the performance of the vehicle, the past driving content is learned in this case, too, and the threshold is set smaller when the left and right shake is large.
[0040]
【The invention's effect】
As described above, according to the lane departure warning device of the present invention, the deviation d from the vehicle line can be obtained by projecting the position of the vehicle obtained by the autonomous navigation onto the line where the road markers are arranged. Based on this deviation d, it is possible to accurately determine and warn of a lane departure.
[0041]
In particular, according to the invention described in claim 2, the position of the vehicle obtained by autonomous navigation can be accurately corrected based on the coordinate data string of the road marker stored in the road marker map memory. It is possible to accurately obtain the deviation d from the line.
[Brief description of the drawings]
FIG. 1 is a view of a road in which a magnetic nail is embedded in the middle of a lane as a road marker, viewed from a low sky.
FIG. 2 is a block diagram showing an in-vehicle device including a lane departure warning device.
FIG. 3 is a diagram showing a circuit configuration in which a plurality of magnetoresistive elements are arranged on an elongated rubber-like permanent magnet and an output voltage is taken out from each of them.
FIG. 4 shows a vehicle estimated position (x _s ) between a vehicle estimated position (x _n−1 , y _n−1 ) at the beginning and a vehicle estimated position (x _n , y _n ) at the end of any detection period Δt. , Y _s ) is a diagram for explaining a calculation method.
[5] the vehicle estimated position (x _s, y _s) and magnetic nails coordinates (x _m, y _m) is a diagram illustrating the positional relationship between the.
FIG. 6 is a diagram for explaining a method of projecting a vehicle estimated position (x, y) to a position on a line where magnetic nails are arranged.
FIG. 7 is a diagram illustrating a method of projecting a vehicle estimated position (x, y) in consideration of a yaw angle φ when projecting to a position on a line where magnetic nails are arranged.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 Magnetoresistance element 2 Resistance element 3 Permanent magnet 11 Traveling distance detection part 12 Traveling direction detection part 13 Magnetic field detection part 13a Waveform shaping part 14 Position estimation part 15 Road marker map memory 16 Position correction part 17 Antenna 17a Reception demodulation part 17b Loop coil 17c Loop coil 17d 90 ° phase shifter 17e Hybrid

Claims (3)

A device that warns a deviation from a lane by detecting a deviation of the vehicle from the road marker by laying a road marker on the road and detecting the road marker,
Vehicle position estimation means for estimating the position of the vehicle by autonomous navigation;
A road marker map memory storing a coordinate data string of road markers,
Deviation detecting means for detecting a lane departure by calculating a deviation d from the vehicle line by projecting the position of the vehicle onto a line on which road markers are arranged, and comparing this deviation d with a threshold value d ₀ , A lane departure warning device comprising a warning means. Marker detection means for detecting a road marker;
The road marker detected by the marker detection unit is associated with the road marker stored in the coordinate data memory, and detected by the marker detection unit based on the vehicle position estimated by the vehicle position estimation unit. The estimated position of the road marker is calculated, and the estimated position of the road marker is compared with the coordinate data of the road marker stored in the coordinate data memory, thereby correcting the position of the vehicle estimated by autonomous navigation. The lane departure warning device according to claim 1, further comprising position correction means. The threshold d _0, the following (a) ~ lane departure warning system according to claim 1, characterized in that to be set according to the criteria of either or a combination thereof (d).
(a) the width of the vehicle and to reduce the threshold d ₀ broad.
(b) the lane width and to reduce the threshold d ₀ narrow.
(c) When the yaw angle φ, which is the difference between the traveling direction of the vehicle and the direction of the line where the magnetic nails are arranged, is large, the threshold d ₀ is decreased.
(d) the distance from the white line defining a lane to guardrail or sidewall and to reduce the threshold d ₀ short.

Images