JP3685991B2 - Signal measuring method and position measuring method having noise removal function - Google Patents

Description

と表される。測定対象物理量は信号ベクトルｖ_s に雑音ベクトルｖ_n から構成されており、その合成ベクトルを測定ベクトルｖ_m と呼ぶことにすれば、
【数２】

である。
【００２２】
同一の地点で得た測定ベクトルに本発明を適用する場合には、信号ベクトルｖ_s に雑音ベクトルｖ_n が平行でも反平行でもないとする。つまり、適当な定数Ｋを用いて
【数３】

とすることはできないとする。
【００２３】
雑音ベクトルの異なる２成分、例えば、第ｊ成分に対する第ｋ成分の成分比をｒとするとき、次の演算を行う。
【数４】

定義により、
【数５】

である。ここでは、下付添字の…，ｋ等は当該ベクトルのｋ番目の成分を指している。
演算結果である参照信号ｆ_s (t) は信号ベクトルｖ_s の時間変化を与える時間関数ｆ_s (t) に比例する。
【００２４】
信号ベクトルｖ_s の時間変化を与える時間関数ｆ_s (t) と雑音ベクトルｖ_n の時間変化を与える時間関数ｆ_n (t) が無相関であれば、参照信号ｆ_r (t) と測定ベクトルｖ_m の測定したい成分ｖ_m,i との積をとり、その低周波数成分の平均をとることにより、測定ベクトルｖ_m の測定したい成分ｖ_m,i 中の信号成分の大きさ、つまり信号ベクトルｖ_s の第ｉ成分の大きさｖ_s,i ＜｜ｆ_s (t) ｜＞を得ることができる。ここで、＜Ｐ＞_T は量Ｐの平均時間Ｔの平均を表す。また、｜Ｐ｜は量Ｐの絶対値を表す。
【００２５】
上記の処理は、信号ベクトルｖ_s の時間変化を与える時間関数ｆ_s (t) の遅れ時間０の自己相関を計算していることになるから、信号ベクトルｖ_s の時間変化を変える時間関数ｆ_s (t) と雑音ベクトルｖ_n の時間変化を与える時間関数ｆ_n (t) が無相関という条件が満たされれば、原理的には、信号ベクトルｖ_s の時間変化を与える時間関数ｆ_s (t) と雑ベクトルｖ_n の時間変化を与える時間関数ｆ_n (t) がともに非定常信号であっても構わない。ただし、信号ベクトルｖ_s の時間変化を与える時間関数ｆ_s (t) が非定常の場合には、平均時間内の信号の平均振幅が得られることになる。
【００２６】
信号ベクトルｖ_s の第ｉ成分の大きさｖ_s,i ＜｜ｆ_s (t) ｜＞_T を得るためには、参照信号ｆ_s (t) の振幅を用いる必要がある。これは勿論既知の量である。この手間を省くために、参照信号ｆ_r (t) の振幅を調整して予め決めた値とすることもできる。
【００２７】
以上の説明はベクトルの２個の成分に対して行ったが、ベクトルの成分が３個の場合も全く同様である。ベクトルの成分が３個の場合には、信号ベクトルｖ_s の成分比と雑音ベクトルｖ_n の成分比の違いが大きく、雑音ベクトル根ｖ_n の成分比の絶対値が小さくなるような２個の成分を選ぶことにより、演算の精度を上げることが可能である。
【００２８】
異なる地点で得た２個の測定ベクトルｖ_m1，ｖ_m2
【数６】

に本発明を適用する場合には、測定ベクトルｖ_m1を構成する雑音ベクトルｖ_n1の、例えば、第ｊ成分ｖ_n1,jに対する雑音ベクトルｖ_m2の第ｋ成分ｖ_n2,kの成分比をｒとするとき、次の演算を行う。
【数７】

【００２９】
定義により、
【数８】

である。ここで、下付添字の…，ｋ等は当該ベクトルのｋ番目の成分を指している。また、ｖ_s0,1,j、ｖ_s0,2,kはそれぞれ、測定ベクトル１の方向を与える単位ベクトルの第ｊ成分，測定ベクトル２の方向を与える単位ベクトルの第ｋ成分を表している。
【００３０】
以上の説明は、測定対象物理量をベクトルとして行ったが、ベクトルの次数を１とすればスカラ量にも適用することができることは当然である。
【００３１】
（成分比の誤差の影響）
一般には、上記の成分比ｒには誤差が含まれる。つまり、真の比をｒ₀ に対して、誤差Δ_r が付け加わっている。
【数９】

このとき、参照信号ｆ_r (t) は
【数１０】

となり、信号ベクトルｖ_s の時間変化を与える時間関数ｆ_s (t) 以外に雑音ベクトルｖ_n の時間変化を与える時間関数ｆ_n (t) に比例する成分が含まれることになる。
【００３２】
式（１２）で与えられる誤差を含む参照信号ｆ_r (t) と目的のベクトル成分ｖ_m,i との積をとり、その結果を時間平均する。信号ベクトルｖ_s の時間変化を与える時間関数ｆ_s (t) の相関時間と雑音ベクトルｖ_n の時間変化を与える時間関数ｆ_n (t) の相関時間に比べて平均時間が十分に長ければ、結果として、
【数１１】

が得られる。式（１３）の結果は誤差を含むが、誤差Δｒが十分に小さければ実用上十分な雑音除去効果が得られる。
【００３３】
さらに除去効果を高めるために、本発明では参照信号に以下の処理を施すことが可能である。
【００３４】
（参照信号の矩形波化）
式（１２）のように参照信号ｆ_r (t) が雑音ベクトルｖ_n の時間変化を与える時間関数ｆ_n (t) に比例する成分を含む場合、その影響は参照信号ｆ_r (t) の振幅が変化するための影響と参照信号ｆ_r (t) の位相がずれるための影響とに分けて考えることができる。参照信号ｆ_r (t) を位相を変えずに矩形波化すれば振幅変化による影響を軽減することができる。
【００３５】
参照信号ｆ_r (t) を矩形波化するためには、元の参照信号ｆ_r (t) を十分に増幅して、適当なレベルで振幅制限するか、又は参照信号ｆ_r (t) の平均値を閾値として、信号レベルが閾値より高い場合には所定の大きさの正値を与え、閾値より低い場合には同一の大きさの負値を与える方法を用いることができる。
何れの方法も数式的には、
【数１２】

なる演算を行うことと等価である。ここで、ａは定数であり、ｆ_r,rect(t) は矩形波化された参照信号である。
【００３６】
（参照信号の周期の調整）
参照信号ｆ_r (t) が雑音ベクトルｖ_n の時間変化を与える時間関数ｆ_n (t) に比例する成分を含む場合、参照信号ｆ_r (t) を単に矩形波化しても、参照信号ｆ_r (t) の位相がずれる影響が残る場合がある。
【００３７】
信号ベクトルｖ_s の時間変化を与える時間関数ｆ_s (t) が周期信号である場合、該矩形波化した参照信号ｆ_r (t) を予め定めた所定の長さの時間区間に分割し、各時間区間内で参照信号ｆ_r (t) の周期が一定になるように参照信号ｆ_r (t) の周期を調整することにより位相がずれる影響を軽減することができる。このためには、検出手段の出力信号をデジタル値化して処理することが有効である。
【００３８】
例えば、矩形波化参照信号ｆ_r,rect(t) を所定の長さの時間区間 [ｔ_k ，ｔ_k+1], ｔ_k+1 −ｔ_k ＝Δｔ_k として、時間区間 [ｔ_k ，ｔ_k+1]内での矩形波化参照信号ｆ_r,rect(t) の周期を平均化することである。ここで、時間ｔ_k ，ｔ_k+1 は時間区間の境界を与える。また、時間区間[ ｔ_k ，ｔ_k+1]の長さΔｔ_k は、矩形波化参照信号ｆ_r,rect(t) 正の期間と負の期間の個数が同時間区間内で整数値になるように決める。
【００３９】
また、矩形波化参照信号ｆ_r,rect(t) の各周期を当該周期を含む予め定めた所定の個数の周期の平均となるように矩形波化参照信号ｆ_r,rect(t) の周期を調整することによっても同様な効果が得られる。
【００４０】
つまり、矩形波化参照信号ｆ_r,rect(t) のｋ番目の周期をＰ_k とするとき、周期Ｐ_k の前後にそれぞれＮ_P ，Ｎ_s 個の周期をとり、合計Ｎ_P ＋Ｎ_s ＋１個の周期の平均周期長＜Ｔ_k ＞を周期Ｐ_k の周期長とすることである。すなわち、
【数１３】

とすることである。ただし、Ｔ_j はｊ番目の周期の周期長である。ここでいう周期とは便宜的に矩形波化参照信号ｆ_r,rect(t) の連続する１個の正の期間と負の期間の組と考える。実際には、勿論、連続する負の期間と正の期間の組であっても良いし、正の期間と負の期間を同等に扱って半周期を単位としても構わない。
【００４１】
（雑音ベクトルｖ_n の成分比の求め方）
本発明では特に以下の２種類の求め方が有効である。
第１の方法は、信号ベクトルｖ_s が存在しない状態で、雑音ベクトルｖ_n の成分比ｒを求める。
例えば、２個の測定ベクトルｖ_m1，ｖ_m2に対して測定ベクトルｖ_m1の第ｊ成分に対する測定ベクトルｖ_m2の第ｋ成分の成分比を求める場合、各成分の絶対値の平均から比を求める。
【数１４】

なる演算を行う。両者は信号成分が無いので同じ量を表す。ここで、雑音ベクトルｖ_m1の第ｊ成分ｖ_n1,jに対する平均時間Ｔ_j と雑音ベクトルｖ_m2の第ｋ成分ｖ_n2,kに対する平均時間Ｔ_k は一般に異なっても構わない。
【００４２】
あるいは、各成分の２乗の平均値の比の平方根を取る演算
【数１５】

による方法が可能である。勿論、測定ベクトルｖ_m1とｖ_m2が同一のものであっても構わない。ただし、この場合ｊ≠ｋであることが必要なのは当然である。
【００４３】
第２の方法は、雑音ベクトルｖ_n の時間変化を与える時間関数ｆ_n (t) が、信号ベクトルｖ_s の時間変化を与える時間関数ｆ_s (t) がほとんど周波数成分を持たないような周波数帯域 [ｆ_r1，ｆ_r2],ｆ_r2−ｆ_r1＝Δｆ_ratio に、周波数成分を持つときに適用する方法である。
【００４４】
成分比ｒを求める測定ベクトルｖ_m1とｖ_m2の２個のベクトル成分ｖ_m1,jとｖ_{m 2,k} とから周波数帯域 [ｆ_r1，ｆ_r2] に含まれる周波数成分ｕ_m1,jとｕ_m2,kを抜き出し、式（１６）あるいは（１７）のｖ_n1,jとｖ_n2,kの代わりに用いる。すなわち、
【数１６】

とすればよい。
【００４５】
上記の式（１８）および（１９）によって成分比を求める手続きの流れを図１および図２に示す。式（１６）および（１７）によって成分比を求めるための手続きの流れは、図１および図２においてフィルタを除いた流れをベクトル成分ｖ_m1,jとｖ_m2,kに適用すればよい。
【００４６】
なお、フィルタの通過周波数帯域に関しては、次の如き構成が可能である。
（１）信号が周波数成分を実質的に持たない帯域に固定する。
（２）測定の環境、すなわち、雑音の状況に応じて中心周波数と帯域幅のいずれかあるいは両方を適宜変えられるようにする。
（３）雑音成分が最大となる周波数帯域になるように動的に変える機能を持たせる。
【００４７】
（信号測定処理の流れ）
本発明による信号測定のための処理の流れの１例を図３と図４に示す。図中、ＰＱＭ−１，ＰＱＭ−２，…，ＰＱＭ−Ｍは測定対象物理量、ＱＤＭ−１，ＱＤＭ−２，…，ＱＤＭ−Ｍは測定対象物理量を検出するための検出手段、ＲＭＰは式（１６）あるいは（１７）によって成分比ＮＣＲを求める成分比測定手順、ＰＲＤ−０，ＰＲＤ−３，ＰＲＤ−４，…，ＰＲＤ−Ｍは２信号の積の演算を行う乗算手段、ＡＤＤは２信号の和の演算を行う加算手段である。ＡＶＲ−３，ＡＶＲ−４，…，ＡＶＲ−Ｍは低域フィルタリング、平均処理あるいはその両方を行う平均化手段である。ＷＦＴは参照信号ＲＥＦの振幅の調節，矩形波化などを行う波形整形手段である。
【００４８】
図３は信号がない状態で予め成分比を求めておく方法を示すブロック図である。図中の波線より上に示されている処理が、成分比ＮＣＲを求める処理の流れである。測定対象物理量−１（ＰＱＭ−１）、測定対象物理量−２（ＰＱＭ−２）をそれぞれ検出手段−１（ＱＤＭ−１）、検出手段−２（ＱＤＭ−２）に入力して検出し、成分比測定手順ＲＭＰにより成分比を求める。
【００４９】
信号を測定するためには、この成分比ＮＣＲを検出手段−２（ＱＤＭ−２）の出力に乗算手段ＰＲＤ−０で乗じて検出手段−１（ＱＤＭ−１）の出力から加算手段ＡＤＤにより減ずる。加算手段ＡＤＤの出力を波形整形手段ＷＥＴで整形して参照信号ＲＥＦとする。目的とする測定対象物理量である測定対象物理量−３（ＰＱＭ−３）、…測定対象物理量−Ｍ（ＰＱＭ−Ｍ）をそれぞれ検出手段−３（ＱＤＭ−３）、…検出手段−Ｍ ＱＤＭ−Ｍで検出し、その出力に乗算手段ＰＲＤ−３、…乗算手段ＰＲＤ−Ｍで参照信号ＲＥＦを乗じ、乗算手段ＰＲＤ−３、…乗算手段ＰＲＤ−Ｍの出力に平均化手段ＡＶＲ−３、…ＡＶＲ−Ｍで平均処理すれば、目的の信号が得られる。
ここで、成分比ＮＣＲを得るために使用している測定対象物理量−１（ＰＱＭ−１）、測定対象物理量−２（ＰＱＭ−２）が目的とする測定対象物理量−３（ＰＱＭ−３）、…測定対象物理量−Ｍ（ＰＱＭ−Ｍ）と同じであっても構わない。
【００５０】
図４では、予め成分比ＮＣＲを求めておくのではなく、信号成分が実質的に存在しない周波数帯域に雑音成分が周波数成分を持つことを利用して、信号の測定を行うと同時に成分比ＮＣＲを求めている。勿論、成分比ＮＣＲの測定と信号の測定を時間的に分けて行っても構わない。図中、ＢＰＦ−１，ＢＰＦ−２は帯域フィルタである。図３の処理との本質的な違いは、検出手段−１（ＱＤＭ−１、検出手段−２（ＱＤＭ−２）の出力に帯域フィルタ−１（ＢＰＦ−１）、帯域フィルタ−２（ＢＰＦ−２）を配置して、測定対象物理量−１（ＰＱＭ−１）、測定対象物理量−２（ＰＱＭ−２）に含まれる雑音成分の当該周波数帯域の成分を取り出していることである。
【００５１】
成分比ＮＣＲの測定は、信号の測定と同時に行うことも別の時間に行うことも可能であることは前述の通りである。さらに、図４の手順によれば、成分比ＮＣＲが信号の測定に要する時間に比べて十分に緩やかに時間的に変化する場合にも、成分比測定手順ＲＭＰで行う平均処理の平均時間を適当に選ぶことにより雑音を有効にかつ効率的に除去することが可能である。
【００５２】
（磁界測定を利用する位置測定方法の基本事項）
本発明の実施例を説明するための参考として、磁界源を利用する位置測定方法に関する基本的な事項について説明する。
磁界を利用して位置を測定する方法は、原理的には同じであるが、形態的に磁界発生源の位置を求める方法と磁界を検出する磁界検出手段の位置を求める方法とに大別される。
【００５３】
（磁界発生源の位置を求める方法）
磁界発生源の位置を求める方法では、磁界発生源は単一で小さな形状である場合がほとんどである。この場合、磁界発生源の位置と姿勢を同時に求めるためには６個の未知数を決める必要があり、位置または方向が異なる最低６個の磁界の大きさが必要である。勿論この方法では、磁界を検出する６個の磁界検出手段の位置と姿勢が既知であることが前提である。
なお、ここでいう位置とは、例えば大地に固定された適当な座標系でのｘ座標，ｙ座標，ｚ座標である。また、姿勢とは同座標系でのｘ軸，ｙ軸，ｚ軸に関する回転角である。
【００５４】
別の方法で磁界発生源の姿勢の情報の一部が分かっていれば、その分だけ測定する磁界の個数を減らすことができる。磁界発生源の姿勢を知る方法としては、地球の重力方向を参照する方法がある。この方法では磁界発生源の鉛直方向に対する傾き、すなわち鉛直方向に直交し、互いに直交する２軸の周りの回転角を知ることができるので、前記３個の回転角の内の２個の値を知ることが可能である。しかし、鉛直方向に関する回転角を知ることはできない。
鉛直方向を参照して２個の回転角を知ることができる場合、測定磁界は最低４個でよい。
【００５５】
原理的には、磁界発生源がどのような空間的な分布の磁界を発生するのかが既知であり、相対的な位置と相対的な姿勢の関係は少なくとも一部が既知でれば複数個の磁界発生源であっても構わない。例えば、磁界発生源が２個であると、１個の磁界検出手段で得られる磁界の情報は２倍になるとともに、磁界発生源に関する未知数も２倍の１２個になる。相対的な位置と相対的な姿勢の関係が全く未知である場合には、１２個の磁界の大きさ、すなわち６個の磁界測定手段が必要となり、個々の磁界発生源を測定するのと変わらないことになる。２個の磁界発生源の相対的な位置が既知であれば、未知数は９個になるので、最低５個の磁界測定手段で９個の磁界を測定すればよい。また、２個の磁界発生源の相対的な位置と姿勢が全て既知であれば、未知数は６個になるので、最低３個の磁界測定手段で６個の磁界を測定すればよい。
【００５６】
（磁界検出手段の位置を求める方法）：
磁界検出手段の位置を求める場合には、磁界検出手段は、例えば、１個の３軸磁気センサのように１個の位置で複数の磁界を検出するものにすることが多い。この場合も未知数は磁界検出手段の位置および姿勢角の６個である。従って、相対的な位置と姿勢が既知である２個の磁界発生源があれば、１個の３軸磁気センサにより６個の磁界を測定してその３軸磁気センサの位置と姿勢角を決めることが可能である。ここで、３軸磁気センサとは互いに直交する３方向の磁界を空間的にほぼ同一の位置で検出することができるようなセンサである。
【００５７】
鉛直方向を基準として３軸磁気センサの姿勢角の内の２個を測定する場合にも、求めるべき未知数は４個になるが、磁界発生源が発生する磁界だけから３軸磁気センサの位置を得るためには２個の磁界発生源が必要になる。ただし、磁界発生源を１個だけとして、何らかの仮定あるいは地磁気の方向から残りの姿勢角を求めて、３個の磁界から位置を得る方法もある。
【００５８】
逆に相対的な位置関係が既知の６個の磁界検出手段と１個の磁界発生源を用いることも可能である。さらに、鉛直方向を基準としてこれらの磁界検出手段群の２個の姿勢角を測定する場合には、磁界検出手段の個数を４個とすることが可能である。
【００５９】
以上の何れの方法においても、磁界発生源や磁界検出手段の個数は多くても構わない。そのときには例えば最小２乗法のような最適解を求める手法により位置を決めればよい。また、これらの位置測定法では交流磁界と静磁界の何れかを用いることもできるが、静磁界を用いる場合には磁界発生源から人為的に発生させた磁界と地磁気を区別するための手段が必要である。この方法としては通常、人為的に発生する磁界の方向を反転させて２回磁界を測定し、２個の測定結果の差分をとる方法が一般的である。
【００６０】
（位置測定法に関する実施例１）
前項（磁界検出手段の位置を求める方法）で説明した、位置測定法の内で交流磁界を用いる方法の何れに対しても本発明の信号測定法を適用することが可能である。すなわち、１個以上の磁界発生源により人為的に発生した交流磁界を１個以上の磁界検出手段で検知して、検知した磁界の大きさおよび磁界検出手段の位置の少なくとも一方を変えたときのその大きさの変化から、磁界発生源の位置もしくは磁界検出手段の位置を算出又は探知する位置測定法において、信号成分に対する雑音成分の大きさの比が異なる２個以上の磁界成分を検知するように該磁界検出手段を配置し、磁界検出手段で検知した測定量に対して本発明の信号測定方法を用いることにより、雑音除去機能を有する位置測定方法とすることが可能である。
【００６１】
本発明による雑音除去機能を有する位置測定方法の一般的な構成の例を図５に示す。図中、ＭＦＳ−１，ＭＦＳ−２，ＭＦＳ−３，…，ＭＦＳ−ｋ，…，ＭＦＳ−ｎは磁界発生源（磁界発生源−１，磁界発生源−２，磁界発生源−３，磁界発生源−ｋ，…，磁界発生源−ｎ）、ＭＦＤ−１，ＭＦＤ−２，ＭＦＤ−３，…，ＭＦＤ−ｊ，…，ＭＦＳ−ｍは磁界検出手段（磁界検出手段−１，磁界検出手段−２，磁界検出手段−３，磁界検出手段−ｊ，…，磁界検出手段−ｍ）、ＱＰＳ−１，ＱＰＳ−２，ＱＰＳ−３，…，ＱＰＳ−ｊ，…，ＱＰＳ−ｍは測定量（測定量−１，測定量−２，測定量−３，…，測定量−ｊ，…，測定量−ｍ）、ＡＤＭ−３，…，ＡＤＭ−ｊ，ＡＤＭ−ｍは姿勢検出手段（姿勢検出手段−３，…，姿勢検出手段−ｊ，…，姿勢検出手段−ｍ）、ＰＬＤ−３，…，ＰＬＤ−ｊ，…，ＰＬＤ−ｍは同期検手段（同期検波手段−３，…，同期検波手段−ｊ，…，同期検波手段−ｍ）、ＲＳＧは参照信号発生手段、ＰＣＭは位置計算手段である。
【００６２】
磁界発生源ＭＦＳ−１，ＭＦＳ−２，ＭＦＳ−３，…，ＭＦＳ−ｋ，…，ＭＦＳ−ｎは、例えば交流電流が流れるコイル，電線，ループ状の電線や機械的に回転する磁石などであり、所定の周波数の交流磁界を発生する。この交流磁界は情報を載せるための振幅変調、位相変調、周波数変調などを受けてもよい。磁界検出手段ＭＦＤ−１，ＭＦＤ−２，ＭＦＤ−３，…，ＭＦＤ−ｊ，…，ＭＦＤ−ｍは空芯あるいはコアのあるコイルやフラックスゲート磁束計など交流磁界を検知して検知した磁界の大きさに比例する信号又は情報を出力することができるものであれば何でもよい。姿勢検出手段ＡＤＭ−３，…，ＡＤＭ−ｊ，…，ＡＤＭ−ｍは加速度センサ，傾斜センサ，回転センサ，ジャイロなどであり、それぞれ磁界検出手段ＭＦＤ−３，…，ＭＦＤ−ｊ，…，ＭＦＤ−ｍと所定の姿勢で固定されており、これら磁界検出手段の姿勢を検出する。位置計算手段ＰＣＭは同期検波手段ＰＬＤ−３，…，ＰＬＤ−ｊ，…，ＰＬＤ−ｍにより得られた信号磁界の大きさと姿勢検出手段ＡＤＭ−３，…，ＡＤＭ−ｊ，…，ＡＤＭ−ｍにより得られた磁界検出手段ＭＦＤ−１，ＭＦＤ−２，ＭＦＤ−３，…，ＭＦＤ−ｊ，…，ＭＦＤ−ｍの姿勢角から、磁界発生源ＭＦＳ−１，ＭＦＳ−２，ＭＦＳ−３，…，ＭＦＳ−ｋ，…，ＭＦＳ−ｎ又は磁界検出手段ＭＦＤ−１，ＭＦＤ−２，ＭＦＤ−３，…，ＭＦＤ−ｊ，…，ＭＦＤ−ｍの位置を算出する。ＮＦＳは雑音磁界発生源であり、本発明の目的にかかわるが、構成要件ではない。
【００６３】
参照信号発生手段ＲＳＧは、図６および図７に構成例を示すようなもので、信号対雑音比が異なる２個の測定量ＰＱＳ−１（測定量−１）およびＰＱＭ−２（測定量−２）を入力として、信号成分の時間変化に比例する参照信号ＲＥＦを作り出す。
【００６４】
図６の構成で、参照信号発生手段ＲＳＧは破線で囲まれた部分である。磁界である測定対象物理量ＰＱＭ−１（測定対象物理量−１）およびＰＱＭ−２（測定対象物理量−２）がそれぞれ磁界検出手段ＭＦＤ−１，ＭＦＤ−２で測定量ＰＱＳ−１，ＰＱＳ−２に変換される。切り替えスイッチＭＳＷは成分比測定と参照信号発生を切り替えるもので、図６は成分比測定の場合の構成である。参照信号を発生する場合には２個のスイッチを下側の端子に、すわなち、測定量を乗算手段に伝えるように切り替える。
【００６５】
成分比測定手段ＲＭＭは式（１６）または（１７）に従い、成分比を求める処理をするもので、ハードウェアで構成されたものでも、ソフトウェアにより構成されたもの、又は両者を混合して用いるものでも構わない。求められた成分比の値はハードウェア的あるいはソフトウェア的に記憶され、参照信号ＲＥＦの発生に用いられる。
【００６６】
参照信号ＲＥＦの発生は、測定量−２（ＰＱＳ−２）に乗算手段ＰＲＤで成分比ｒを乗じて加算手段ＡＤＤにより測定量−１（ＰＱＳ−１）より減ずることにより、測定量中の信号成分と同相の時間変化をする信号を作り出すことによって行う。また、必要に応じて、波形整形手段ＷＦＴで矩形波化などの処理を施し参照信号ＲＥＦとする。
【００６７】
図７の構成では、参照信号ＲＥＦの発生と成分比ｒの測定を同時に行うことが可能である。図６と同様に破線で囲まれた内部が参照信号発生手段ＲＳＧである。参照信号ＲＥＦの発生と成分比ｒの測定を同時に行うために、帯域フィルタＢＰＦ−１（帯域フィルタ−１）、ＢＰＦ−２（帯域フィルタ−２）により、それぞれ測定量ＰＱＳ−１，ＰＱＳ−２中の特定の周波数帯域の成分を取り出す。この周波数帯域の決め方に関しては、次のように処理することができる。
（１）信号が周波数成分実質的に持たない帯域に固定する。
（２）測定の環境、すなわち、雑音の状況に応じて中心周波数と帯域幅のいずれかあるいは両方を適宜変えられるようにする。
（３）雑音成分が最大となる周波数帯域になるように動的に変える機能をもたせる。
【００６８】
図６および図７の参照信号発生手段ＲＳＧの構成要素はソフトウェアで構成されたもの、ハードウェアで構成されたもの、両者が混在したものの何れであっても構わない。
同期検波手段ＰＬＤ−３，…，ＰＬＤ−ｊ，…，ＰＬＤ−ｍは、図８に１構成例を示すようなもので、測定量ＰＱＳ−１，ＰＱＳ−２，ＰＱＳ−３，…，ＰＱＳ−ｊ，…，ＰＱＳ−ｍに参照信号ＲＥＦを乗じて、低域フィルタリング、平均等の処理を行い、これら測定量中の参照信号ＲＥＦと同相の信号検出をする。
【００６９】
図８は同期検波手段ＰＬＤの１構成例である。同期検波手段ＰＬＤは破線に囲まれた部分である。図において、磁界である測定対象物理量ＰＱＭを磁界検出手段ＭＦＤで検出して測定量ＰＱＳを求める。乗算手段ＰＲＤを用いて参照信号ＲＥＦを測定量ＰＱＳに乗じた磁界の低周波数成分を低域フィルタＬＰＦで取りだし、積分器ＩＴＧで平均して信号ＳＩＧを求める。ここで、低域フィルタＬＰＦは必ずしも必要でない。図８中の同期検波手段ＰＬＤの構成要素はソフトウェアで構成されたもの、ハードウェアで構成されたもの、両者が混在したものの何れであっても構わない。
【００７０】
（位置測定法に関する実施例２）
図９は本発明による雑音除去機能を有する位置測定装置の実施例の構成であり、磁界発生源の位置を測定するものである。
ＭＦＳは磁界発生源、ＡＤＭは姿勢検出手段、ＦＲＭはフレームである。本実施例では、磁界検出手段ＭＦＤ−１，ＭＦＤ−２，ＭＦＤ−３、姿勢検出手段ＡＤＭ、参照信号発生手段ＲＳＧ、同期検波手段ＰＬＤ−１，ＰＬＤ−２，ＰＬＤ−３、位置計算手段ＰＣＭは何れもフレームＦＲＭに固定されている。磁界発生源ＭＦＳの位置の測定は、フレームＦＲＭを移動して、磁界発生源ＭＦＳが発生する磁界が特定の条件を満足する場所、例えば、ある水平面内で特定の方向の磁界強度が最大になる場所を探索し、その位置とそこでの磁界強度から磁界発生源ＭＦＳの位置を参照あるいは推定する。
【００７１】
磁界検出手段−１（ＭＦＤ−１）と磁界検出手段−２（ＭＦＤ−２）は参照信号を発生するためと位置測定の両方に使用されているから、参照信号発生手段は図７のような構成のものであることが必要である。
姿勢検出手段ＡＤＭは必ずしもハードウェアとして備えている必要はない。フレームＦＲＭを移動する測定者がフレームＲＦＭの姿勢を所定の方向に保つことにより同等の機能を果たしても良い。また、磁界検出手段ＭＦＤは最低２個で十分である。この場合、磁界検出手段−３（ＭＦＤ−３）を除いたものとなり、構成要素数が少ない利点がある。
【００７２】
（位置測定法に関する実施例３）
図１０は空間的に離れた複数個の位置に磁界検出手段を設置して磁界発生源ＭＦＳの位置を測定する本発明の実施例である。図１０は空間的に離れた３箇所に磁界検出手段をおく場合の構成である。
図中、ＵＭＭ−１，ＵＭＭ−２，ＵＭＭ−３は磁界検出手段として用いられる単位測定手段である。ＰＭＭは位置測定手段であり、信号受信手段ＳＲＸと位置計算手段ＰＣＭとにより構成されている。本実施例では磁界検出手段を空間的に離れた複数の位置に置くために、信号送出機能を有する単位測定手段ＵＭＭ−１（単位測定手段−１），ＵＭＭ−２（単位測定手段−２）、ＵＭＭ−３（単位測定手段−３）を用いている。これにより、位置測定手段ＰＭＭ中の信号受信手段ＳＲＸとの間で信号の授受を行う。
【００７３】
単位測定手段ＵＭＭの構成例を図１１および図１２に示す。図１１の例では単位測定手段ＵＭＭは、磁界検出手段ＭＦＤ−１，ＭＦＤ−２，ＭＦＤ−３、姿勢検出手段ＡＤＭ、参照信号発生手段ＲＳＧ、同期検波手段ＰＬＤ−１，ＰＬＤ−２，ＰＬＤ−３、信号送出手段ＳＴＸから構成されている。磁界検出手段ＭＦＤ−１，ＭＦＤ−２，ＭＦＤ−３は独立な磁界を検出することが可能であれば、必ずしも互いに直交する方向の磁界を検出するように配置する必要はないが、典型的には本発明のように磁界検出手段ＭＦＤ−１，ＭＦＤ−２，ＭＦＤ−３は互いに直交する異なる３方向の磁界強度を実質的に同一の地点で検出する３軸磁気センサを構成する。姿勢検出手段ＡＤＭは、磁界検出手段ＭＦＤ−１，ＭＦＤ−２，ＭＦＤ−３と固定されており、これらの姿勢角を検出する。信号送出手段ＳＴＸは、同期検波手段ＰＬＤ−１，ＰＬＤ−２，ＰＬＤ−３の出力および姿勢検出手段ＡＤＭの出力を電磁界信号，電圧信号，電流信号，あるいは光信号として送出する機能を持つ。
【００７４】
図１２は直交する２つの方向の磁界を検出するための単位測定手段ＵＭＭである。そのために、図１１の構成に比べて磁界検出手段ＭＦＤ−３および同期検波手段ＰＬＤ−3 が無い。
【００７５】
図１０中の信号受信手段ＳＲＸは単位測定手段ＵＭＭ中の信号送出手段ＳＴＸから送られてくる同期検波手段ＰＬＤ−１，ＰＬＤ−２，ＰＬＤ−３の出力（図１１の構成の場合）あるいは同期検波手段ＰＬＤ−１，ＰＬＤ−２の出力（図１２の構成の場合）および姿勢検出手段ＡＤＭの出力を受信して位置計算手段ＰＣＭに渡す。
【００７６】
図１３は、単位測定手段が単位測定手段−１（ＵＭＭ−１）、単位測定手段−２（ＵＭＭ−２）の２個の場合の実施例である。
【００７７】
独立な６個の磁界の大きさが必要な場合には、図１０の実施例で図１２の構成の単位測定手段ＵＭＭを使用するか、図１３の実施例で図１１の構成の単位測定手段ＵＭＭを使用するかすればよい。勿論、図１０の構成で図１１の構成の単位測定手段ＵＭＭを使用しても全く支障はない。
【００７８】
さらに、図１０および図１３の実施例では、図１４の構成の単位測定手段ＵＭＭを図１１の構成の単位測定手段ＵＭＭの代わりに用いることができる。
【００７９】
図１４の単位測定手段ＵＭＭでは、２個の磁界検出手段，磁界検出手段−１（ＭＦＤ−１）及び時間検出手段−２（ＭＦＤ−２）は空間的に別の場所に置かれている。これらの磁界検出手段の姿勢を検知するための姿勢検出手段、姿勢検出手段−１（ＡＤＭ−１），姿勢検出手段−２（ＡＤＭ−２）が、それぞれ磁界検出手段−１（ＭＦＤ−１），磁界検出手段−２（ＭＦＤ−２）に固定されている。磁界検出手段−１（ＭＦＤ−１）の出力信号と姿勢検出手段−１（ＡＤＭ−１）の出力信号、および、磁界検出手段−２（ＭＦＤ−２）の出力信号と姿勢検出手段−２（ＡＤＭ−２）の出力信号は、それぞれ検出信号処理手段ＳＰＭに入力され、図１１および図１２の場合と同様な処理を受ける。
【００８０】
（位置測定法に関する実施例４）
図１５は磁界発生源の２個の姿勢角に関する情報を利用することができるために、独立な４個の磁界の大きさが必要な場合の実施例である。この場合、図１２の構成の単位測定手段ＵＭＭを用いればよい。勿論、図１１の構成の単位測定手段ＵＭＭを用いても支障はない。
【００８１】
（位置測定法に関する実施例５）
図１６は、相互の位置および姿勢が既知である２個の磁界発生源を用いて磁界検出手段の位置を測定する場合の測定方法を実施するための構成である。
磁界発生源−１（ＭＦＳ−１）と磁界発生源−２（ＭＦＳ−２）は、
（１）周波数が異なる磁界を発生する、
（２）異なる符号でコーディングされた磁界を発生する、
（３）異なる時間に磁界を発生する、
などの方法により、それぞれの磁界発生源が発生する磁界を識別することができるような仕方で交流磁界を発生する、
磁界検出手段ＭＦＤ−１，ＭＦＤ−２，ＭＦＤ−３は互いに直交する異なる３方向の磁界強度を実質的に同一の地点で検出する３軸磁気センサを構成する。
【００８２】
（位置測定法に関する実施例６）
図１７は、相互の位置および姿勢が既知である２個の磁界発生源を用いて磁界検出手段の位置を測定する場合の測定法の別の構成である。本実施例では磁界強度を測定する部分と、位置を計算する部分を空間的に離れた場所に置いている。
【００８３】
【発明の効果】
以上詳細に説明したように、本発明によれば、磁界や電界などを媒介として物理量の測定を行う場合の雑音の除去に広く適用可能であり、地下埋設電力線などのように、測定に影響を与える雑音電流が近くにある場合にも、精度の高い位置測定が可能となり、水平ドリリング等を磁気的雑音が多い都市部で行う場合に掘削位置の測定を行うことが可能となるため、実用的効果は極めて大きい。
【図面の簡単な説明】
【図１】本発明方法に用いる信号ベクトルと雑音ベクトルが存在するときの成分比を求める処理手順を説明するためのブロック図である。
【図２】本発明方法に用いる信号ベクトルと雑音ベクトルが存在するときの成分比を求める処理手順を説明するためのブロック図である。
【図３】本発明方法において雑音のみが存在するときに成分比を求めておく手順を説明するためのブロック図である。
【図４】本発明方法において、信号成分が存在しない周波数帯で雑音を検出して成分比を求める手順を説明するためのブロック図である。
【図５】本発明による雑音除去機能を有する位置測定方法を実施するための構成例を示すブロック図である。
【図６】本発明に用いる参照信号発生手段の１例を示すブロック図である。
【図７】本発明に用いる参照信号発生手段の他の例を示すブロック図である。
【図８】本発明に藻位置いる同期検波手段の構成例を示すブロック図である。
【図９】本発明による雑音除去機能を有する位置測定方法を実施するための他の構成例を示すブロック図である。
【図１０】本発明による雑音除去機能を有する位置測定方法を実施するための他の構成例を示すブロック図である。
【図１１】図１０に示す本発明による雑音除去機能を有する位置測定方法で使用する単位測定手段の１構成例を示すブロック図である。
【図１２】図１０に示す本発明による雑音除去機能を有する位置測定方法で使用する単位測定手段の他の構成例を示すブロック図である。
【図１３】本発明による雑音除去機能を有する位置測定方法を実施するためのさらに他の構成例を示すブロック図である。
【図１４】図１０又は図１３の構成例で使用する単位測定手段の構成例を示すブロック図である。
【図１５】本発明による雑音除去機能を有する位置測定方法を実施するためのさらに他の構成例を示すブロック図である。
【図１６】本発明により磁界検知手段の位置を測定するための構成例を示すブロック図である。
【図１７】本発明により磁界検知手段の位置を測定するための他の構成例を示すブロック図である。
【図１８】従来の生体磁気の測定における雑音除去方法を説明するための略図である。
【符号の説明】
ＰＱＭ−１〜ＰＱＭ−Ｍ 測定対象物理量１〜Ｍ
ＱＤＭ−１〜ＱＤＭ−Ｍ 検出手段１〜Ｍ
ＲＭＰ 成分比測定手順
ＮＣＲ 成分比
ＰＲＤ−０〜ＰＲＤ−Ｍ 乗算手段０〜Ｍ
ＡＤＤ 加算手段
ＷＥＴ 波形整形手段
ＲＥＦ 参照信号
ＡＶＲ−３〜ＡＶＲ−Ｍ 平均化手段−３〜Ｍ
ＳＩＧ−３〜ＳＩＧ−Ｍ 信号−３〜Ｍ
ＢＰＦ−１，ＢＰＦ−２ 帯域フィルタ１，２
ＭＦＳ−１〜ＭＦＳ−ｎ 磁界発生源１〜ｎ
ＭＦＤ−１〜ＭＦＤ−ｍ 磁界検出手段１〜ｍ
ＰＱＳ−１〜ＰＱＳ−ｍ 測定量−１〜ｍ
ＲＳＧ 参照信号発生手段
ＰＬＤ−１〜ＰＬＤ−ｍ 同期検波手段−１〜ｍ
ＰＣＭ 位置計算手段
ＲＭＭ 成分比測定手段
ＬＰＦ 低域フィルタ
ＩＴＧ 積分器
ＦＲＭ フレーム
ＮＦＳ 雑音磁界発生源
ＳＲＸ 信号受信手段
ＰＭＭ 信号測定手段
ＡＤＭ 姿勢検出手段
ＵＭＭ−１〜ＵＭＭ−３ 単位測定手段−１〜３
ＵＭＰ−１〜ＵＭＰ−２ 単位測定手段対１〜２
ＳＰＭ 検出信号処理手段
ＳＴＸ 信号送出手段[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a signal measuring method for measuring a signal component by removing noise when a physical quantity (field) having spatial expansion such as a magnetic field, an electric field, and an electromagnetic field from a generation source includes noise, The present invention also relates to a position measuring method for measuring the position of the source by reducing the influence of noise.
[0002]
[Prior art]
One related technique of this type is a method for reducing external noise in biomagnetism measurement.
In the measurement of biomagnetism, a plurality of magnetic sensors (such as SQUID) are arranged in a magnetically shielded environment as shown in FIG. 17, and a magnetic field generated from a human body is measured. The magnetic field generated by the test body is detected by the magnetic sensor 3 in the figure, but since the external magnetic field generated from other than the test body is simultaneously detected by the magnetic sensor 3, the magnetic field generated by the test body is obtained. Need to remove the component derived from the external magnetic field from the output of the magnetic sensor 3.
[0003]
As a method for this, the magnetic sensor 1 is arranged at a position away from the test body, and the correlation between the outputs of the magnetic sensor 1 and the magnetic sensor 3 when there is no test body is obtained. At this time, the magnetic sensor 1 is separated from the specimen so that the magnetic field generated by the specimen (usually very weak) can be ignored. If the component derived from the disturbance magnetic field is removed by subtracting the output of the magnetic sensor 1 based on the correlation previously measured from the output of the magnetic sensor 3, the magnetic field generated from the specimen is measured without being affected by the disturbance. be able to.
[0004]
In another method, in addition to the magnetic sensor 3 for measuring the magnetic field generated from the specimen, the magnetic sensor 1 and the magnetic sensor 2 are separated from the specimen so that the magnetic field generated by the specimen (usually very weak) can be ignored. The spatial gradient of the disturbing magnetic field is obtained from the outputs of the magnetic sensor 1 and the magnetic sensor 2, and the magnitude of the disturbing magnetic field at the position of the magnetic sensor 3 is estimated thereby. There is a way to subtract from.
[0005]
Summarizing the above conventional methods, in any method, the first detection means is installed in a place where only the noise signal can be detected, and the noise signal detected by the first detection means and the target signal are measured. The correlation with the noise signal at the place where the second detection means for installing is measured or predicted in advance, and based on the signal detected by the first detection means, the second detection means The noise is removed by a procedure of predicting and subtracting the noise signal included in the output signal.
[0006]
[Problems to be solved by the invention]
Even if the signal characteristics, such as the time change of the signal to be detected, are known, if the noise intensity is larger than the signal intensity, the optimal filter or adaptive filter may not provide sufficient noise removal effects. is there. For example, if the signal is a simple sine wave and an irregular signal with a high-level and wide-band frequency spectrum is superimposed on it, and the irregular signal has a large power in the frequency band of the signal, It is difficult to know the size of the signal accurately with normal filtering.
[0007]
As one method, if there is a reference signal synchronized with the signal, synchronous detection is performed based on the phase of the reference signal, and if the detection output is integrated for a long time, the noise is effective if the noise and the signal are not completely in phase. Can be removed.
However, such a reference signal is not always obtained, and in such a case, it has conventionally been impossible to accurately detect the signal.
[0008]
For example, there are the following examples of such situations.
There is a method called horizontal drilling method as one of the non-open cutting methods. In this method, a small-diameter pipe with a diameter of 100 mm or less is pushed into the ground and drilled, so it is carried out by a normal small-diameter propulsion method. It is not possible to prepare a precise position measuring device near the drilling tip. Therefore, a method is generally employed in which an alternating magnetic field is generated by a coil placed in a drill head, the magnetic field is detected by a magnetic sensor such as a ground coil, and the position thereof is measured.
[0009]
Although this method is simple, since the magnetic field generated by the coil is a dipole magnetic field, it rapidly decays as the distance from the coil increases. If there is a source, there is a disadvantage that reliable measurement becomes difficult. The coil is driven by a battery as a power source, and there is no means for obtaining a reference signal having the same phase as the magnetic field generated by the coil. There may be a method of extracting the reference signal by a signal line passed through the drill pipe. However, in addition to a significant reduction in work efficiency, it is necessary to further extend the signal line to the ground magnetic sensor as the drilling progresses. Yes, it is not practical due to poor workability.
[0010]
A filter for removing specific external noise, which is one of other noise reduction means, is not practical because the statistical properties of the external noise differ from construction site to construction site. Also, most of the external noise cannot be said to be steady enough at the maximum time of about 1 minute that is allowed for one position measurement, but it takes a long time position measurement so that it can be regarded as a steady process, in exchange for a decrease in work efficiency. Unless it is, the application of the adaptive filter is not practical.
[0011]
For these reasons, it has been difficult to realize a position measuring device that is less susceptible to external noise by the horizontal drilling method.
[0012]
The present invention is intended for cases where the following conditions (1) and (2) are satisfied.
(1) Noise exists in addition to the target signal, and noise is included in the output of the detection means.
(2) The detection means cannot be arranged so that only noise can be detected as a signal due to spatial limitations, and the conventional noise removal method cannot be applied. Alternatively, it is possible to arrange the detecting means for detecting only noise as in the conventional method, but it is not practical because the point where the target signal should be measured is not fixed.
[0013]
An object of the present invention is to provide a signal measurement method capable of suppressing noise included in a measurement signal of a physical quantity having a spatial spread from a generation source in consideration of the above-described drawbacks of the conventional technology. is there.
Furthermore, another object of the present invention is to provide a method for measuring the position of a source that is not easily affected by external noise for the non-cutting method under similar circumstances.
[0014]
[Means for Solving the Problems]
In order to achieve this object, the present invention solves the problem by the following method. Although it is the same in principle, the measurement situation is divided into two for the technique used to solve the problem in the present invention.
First, the case where the physical quantity to be measured is measured at two or more points that are spatially different will be described. Each measurement quantity obtained by detecting the physical quantity to be measured by the detection means includes both a signal component and a noise component.
(1) When signal components included in each measurement amount change with time,
(2) When the noise components included in each measurement amount change with time,
(3) When the ratio of the noise and the signal size included in at least two of the measured quantities is different,
In addition, with respect to measurement quantity 1 and measurement quantity 2 which are two measurement quantities having different ratios of noise and signal magnitude, the magnitude of noise contained in measurement quantity 1 with respect to the magnitude of noise contained in measurement quantity 2 And a reference signal is created by subtracting the measurement amount 1 multiplied by the component ratio from the measurement amount 1. The magnitudes of the signals included in measurement quantity 1 and measurement quantity 2 are unknown, but the reference signal changes with time in the same way as the signal. Therefore, if the target measurement is synchronously detected with this reference signal, the magnitude of the signal included in the measurement can be measured.
[0015]
Here, in order to obtain the component ratio, two methods are used. The first method is a method of obtaining the magnitude of the noise component when there is no signal. This method is particularly effective when the position of the detection means for detecting the physical quantity to be measured is spatially fixed. The second method is used when the noise component has a frequency component in a frequency band that does not have the frequency component of the signal line component. The noise component in the frequency band of the measurement amount 1 and the measurement amount 2 is extracted and the component ratio is obtained. Ask for. This method is particularly effective when measurement is performed while moving by a detection means for detecting a physical quantity to be measured.
[0016]
In the above description, the physical quantity to be measured is a scalar, but it may be an arbitrary component of a vector quantity.
[0017]
Next, a case where the measurement target physical quantity is a vector quantity will be described. For convenience of explanation, the vector formed by the signal that is the physical quantity to be measured is the signal vector, the vector that is formed by the noise that is the physical quantity to be measured is the noise vector, and the physical quantity that is formed by superimposing the signal vector and the noise vector is measured. Let's call it a vector.
[0018]
However, in the following description, symbols that use “vector” but not vector symbols are used for all vectors.
[0019]
When the physical quantity to be detected is a vector quantity, each component of the signal vector has the same change in time, and each component of the noise vector has the same change in time, in the substantially same place When the two components of the measurement vector detected by the detection means are the measurement amount 1 and the measurement amount 2, the noise component and the signal component included in the measurement amount 1 and the measurement amount 2 as long as the signal vector and the noise vector are not parallel or antiparallel. The ratio of the sizes of can be different. If the same processing as in the first case is performed on the measurement amount 1 and the measurement amount 2, a signal from which noise has been removed can be obtained.
[0020]
In addition, if the above signal measurement method is applied to the output of the magnetic sensor used in the position measurement method that detects the artificially generated magnetic field and measures the position based on the magnitude, etc., the position measurement is less susceptible to external noise. The law can be realized.
[0021]
DETAILED DESCRIPTION OF THE INVENTION
The physical quantity to be measured is a vector and the signal vector v_sNoise vector v_nAre superimposed. Signal vector v_s, Noise vector v_nBoth can be expressed in the form of unit function time function times. That is, the signal vector v_sThe unit vector in the direction of_s0, A time function giving the time change is f_s(t), noise vector v_nUnit vector v in the direction of_n0, A time function giving the time change is f_n(t)
[Expression 1]

It is expressed. The physical quantity to be measured is the signal vector v_sNoise vector v_nAnd the resultant vector is the measurement vector v_mIf we call it
[Expression 2]

It is.
[0022]
When the present invention is applied to measurement vectors obtained at the same point, the signal vector v_sNoise vector v_nAre neither parallel nor antiparallel. In other words, using an appropriate constant K
[Equation 3]

You can't.
[0023]
When r is a component ratio of two components having different noise vectors, for example, the k-th component to the j-th component, the following calculation is performed.
[Expression 4]

By definition
[Equation 5]

It is. Here, the subscripts ..., k, etc. indicate the k-th component of the vector.
Reference signal f which is the calculation result_s(t) is the signal vector v_sTime function f giving the time change of_sIt is proportional to (t).
[0024]
Signal vector v_sTime function f giving the time change of_s(t) and noise vector v_nTime function f giving the time change of_nIf (t) is uncorrelated, the reference signal f_r(t) and measurement vector v_mComponent to be measured_{m, i}And taking the average of the low frequency components, the measurement vector v_mComponent to be measured_{m, i}The magnitude of the signal component in the signal vector v_sThe size v of the i-th component of_{s, i}<| F_s(t) |> can be obtained. Where <P>_TRepresents the average of the average time T of the quantity P. | P | represents the absolute value of the quantity P.
[0025]
The above processing is performed using the signal vector v_sTime function f giving the time change of_sSince the autocorrelation of the delay time 0 of (t) is calculated, the signal vector v_sTime function f to change the time change of_s(t) and noise vector v_nTime function f giving the time change of_nIn principle, if the condition that (t) is uncorrelated is satisfied, the signal vector v_sTime function f giving the time change of_s(t) and miscellaneous vector v_nTime function f giving the time change of_nBoth (t) may be non-stationary signals. Where the signal vector v_sTime function f giving the time change of_sWhen (t) is non-stationary, the average amplitude of the signal within the average time is obtained.
[0026]
Signal vector v_sThe size v of the i-th component of_{s, i}<| F_s(t) | >>_TIn order to obtain the reference signal f_sIt is necessary to use the amplitude of (t). This is of course a known quantity. In order to save this trouble, the reference signal f_rThe amplitude of (t) can be adjusted to a predetermined value.
[0027]
Although the above description has been made for two components of a vector, the same applies to the case of three vector components. If there are three vector components, the signal vector v_sComponent ratio and noise vector v_nThe difference in the component ratio is large and the noise vector root v_nBy selecting two components that reduce the absolute value of the component ratio, it is possible to improve the calculation accuracy.
[0028]
Two measurement vectors v obtained at different points_m1, V_m2
[Formula 6]

When the present invention is applied to the measurement vector v_m1The noise vector v_n1For example, the j-th component v_{n1, j}Noise vector for_m2K-th component v_{n2, k}When the component ratio of r is r, the following calculation is performed.
[Expression 7]

[0029]
By definition
[Equation 8]

It is. Here, the subscripts ..., k, etc. indicate the k-th component of the vector. And v_{s0,1, j}, V_{s0,2, k}Represents the j-th component of the unit vector that gives the direction of the measurement vector 1, and the k-th component of the unit vector that gives the direction of the measurement vector 2, respectively.
[0030]
In the above description, the physical quantity to be measured is used as a vector. However, if the order of the vector is 1, it can be applied to a scalar quantity.
[0031]
(Influence of component ratio error)
In general, the component ratio r includes an error. In other words, the true ratio is r₀Error Δ_rHas been added.
[Equation 9]

At this time, the reference signal f_r(t) is
[Expression 10]

And the signal vector v_sTime function f giving the time change of_sIn addition to (t), noise vector v_nTime function f giving the time change of_nA component proportional to (t) is included.
[0032]
Reference signal f including the error given by equation (12)_r(t) and the desired vector component v_{m, i}And the results are time averaged. Signal vector v_sTime function f giving the time change of_s(t) correlation time and noise vector v_nTime function f giving the time change of_nIf the average time is sufficiently long compared to the correlation time of (t), the result is
## EQU11 ##

Is obtained. The result of equation (13) includes an error, but if the error Δr is sufficiently small, a practically sufficient noise removal effect can be obtained.
[0033]
In order to further enhance the removal effect, the present invention can perform the following processing on the reference signal.
[0034]
(Rectangular reference signal)
Reference signal f as in equation (12)_r(t) is the noise vector v_nTime function f giving the time change of_nWhen a component proportional to (t) is included, the effect is the reference signal f._r(t) Influence of changing amplitude and reference signal f_rThis can be divided into the effects of the phase shift of (t). Reference signal f_rIf (t) is converted into a rectangular wave without changing the phase, the influence of the amplitude change can be reduced.
[0035]
Reference signal f_rTo convert (t) into a rectangular wave, the original reference signal f_r(t) is sufficiently amplified and amplitude limited at an appropriate level, or the reference signal f_rUsing the average value of (t) as a threshold, a method of giving a positive value of a predetermined magnitude when the signal level is higher than the threshold and giving a negative value of the same magnitude when lower than the threshold can be used. .
Both methods are mathematically
[Expression 12]

Is equivalent to performing the following operation. Where a is a constant and f_{r, rect}(t) is a rectangular reference signal.
[0036]
(Adjustment of reference signal cycle)
Reference signal f_r(t) is the noise vector v_nTime function f giving the time change of_nreference signal f if it contains a component proportional to (t)_rEven if (t) is simply converted into a rectangular wave, the reference signal f_rIn some cases, the effect of the phase shift of (t) may remain.
[0037]
Signal vector v_sTime function f giving the time change of_sWhen (t) is a periodic signal, the rectangular reference signal f_r(t) is divided into time intervals of a predetermined length and a reference signal f within each time interval._rReference signal f so that the period of (t) is constant._rBy adjusting the period of (t), the influence of phase shift can be reduced. For this purpose, it is effective to digitize and process the output signal of the detection means.
[0038]
For example, the rectangular wave reference signal f_{r, rect}(t) is a time interval [t_k, T_{k + 1}], t_{k + 1}-T_k= Δt_kAs the time interval [t_k, T_{k + 1}] The rectangular wave reference signal f_{r, rect}It is to average the period of (t). Where time t_k, T_{k + 1}Gives the boundaries of the time interval. In addition, the time interval [t_k, T_{k + 1}] Length Δt_kIs a rectangular wave reference signal f_{r, rect}(t) The number of positive and negative periods is determined to be an integer value within the same time interval.
[0039]
Further, the rectangular wave reference signal f_{r, rect}The rectangular wave reference signal f is set so that each period of (t) is an average of a predetermined number of periods including the period._{r, rect}The same effect can be obtained by adjusting the period of (t).
[0040]
That is, the rectangular wave reference signal f_{r, rect}The kth period of (t) is P_kThe period P_kN before and after_P, N_sTakes a cycle, total N_P+ N_s+1 average period length <T_k> Cycle P_kThe period length is set as follows. That is,
[Formula 13]

It is to do. T_jIs the cycle length of the jth cycle. The period here is a rectangular reference signal f for convenience._{r, rect}Consider a set of one positive period and a negative period in (t). In practice, of course, it may be a set of consecutive negative periods and positive periods, or the positive period and the negative period may be treated equally and a half cycle may be used as a unit.
[0041]
(Noise vector v_nOf component ratio)
In the present invention, the following two methods are particularly effective.
The first method uses the signal vector v_sIn the absence of noise vector v_nThe component ratio r is obtained.
For example, two measurement vectors v_m1, V_m2Measurement vector v_m1Measurement vector v for the j-th component of_m2When calculating the component ratio of the k-th component, the ratio is determined from the average of the absolute values of the components.
[Expression 14]

Perform the following operation. Both represent the same amount because there is no signal component. Where the noise vector v_m1J-th component v_{n1, j}Average time T for_jAnd noise vector v_m2K-th component v_{n2, k}Average time T for_kMay generally be different.
[0042]
Or the operation which takes the square root of the ratio of the mean value of the square of each component
[Expression 15]

Is possible. Of course, the measurement vector v_m1And v_m2May be the same. However, it is natural that j ≠ k in this case.
[0043]
The second method uses the noise vector v_nTime function f giving the time change of_n(t) is the signal vector v_sTime function f giving the time change of_sThe frequency band [f where (t) has almost no frequency component [f_r1, F_r2], f_r2-F_r1= Δf_ratioThis method is applied when the frequency component is present.
[0044]
Measurement vector v for determining component ratio r_m1And v_m2Two vector components of v_{m1, j}And v_{m 2, k}To the frequency band [f_r1, F_r2] The frequency component u included in_{m1, j}And u_{m2, k}And v in equation (16) or (17)_{n1, j}And v_{n2, k}Use instead of. That is,
[Expression 16]

And it is sufficient.
[0045]
The flow of the procedure for obtaining the component ratio by the above equations (18) and (19) is shown in FIGS. The flow of the procedure for obtaining the component ratio according to the equations (16) and (17) is the same as the vector component v in FIG. 1 and FIG._{m1, j}And v_{m2, k}Apply to the above.
[0046]
The following configuration is possible for the pass frequency band of the filter.
(1) The signal is fixed to a band having substantially no frequency component.
(2) Either or both of the center frequency and the bandwidth can be appropriately changed according to the measurement environment, that is, the noise situation.
(3) Provide a function of dynamically changing the frequency band to maximize the noise component.
[0047]
(Signal measurement process flow)
An example of the flow of processing for signal measurement according to the present invention is shown in FIGS. In the figure, PQM-1, PQM-2, ..., PQM-M are physical quantities to be measured, QDM-1, QDM-2, ..., QDM-M are detection means for detecting the physical quantities to be measured, and RMP is an equation ( 16) or component ratio measurement procedure for obtaining the component ratio NCR according to (17), PRD-0, PRD-3, PRD-4,..., PRD-M are multiplication means for calculating the product of two signals, and ADD is two signals Adding means for calculating the sum of. AVR-3, AVR-4,..., AVR-M are averaging means for performing low-pass filtering, averaging processing, or both. WFT is a waveform shaping means for adjusting the amplitude of the reference signal REF, making a rectangular wave, and the like.
[0048]
FIG. 3 is a block diagram showing a method for obtaining the component ratio in advance in the absence of a signal. The process shown above the wavy line in the figure is the process flow for obtaining the component ratio NCR. The measurement target physical quantity-1 (PQM-1) and the measurement target physical quantity-2 (PQM-2) are input to and detected by the detection means-1 (QDM-1) and detection means-2 (QDM-2), respectively. The component ratio is obtained by the ratio measurement procedure RMP.
[0049]
In order to measure the signal, the component ratio NCR is multiplied by the output of the detecting means-2 (QDM-2) by the multiplying means PRD-0 and subtracted from the output of the detecting means-1 (QDM-1) by the adding means ADD. . The output of the adding means ADD is shaped by the waveform shaping means WET to obtain a reference signal REF. Measurement target physical quantity-3 (PQM-3), which is a target measurement target physical quantity,... Measurement target physical quantity-M (PQM-M) is detected means-3 (QDM-3),... Detection means-M QDM-M , The multiplication means PRD-3,... The multiplication means PRD-M multiplies the reference signal REF, and the multiplication means PRD-3,... The multiplication means PRD-M outputs the averaging means AVR-3,. If average processing is performed at −M, a target signal is obtained.
Here, the measurement target physical quantity-1 (PQM-1), the measurement target physical quantity-2 (PQM-2) used to obtain the component ratio NCR is the measurement target physical quantity-3 (PQM-3), ... It may be the same as the physical quantity to be measured-M (PQM-M).
[0050]
In FIG. 4, the component ratio NCR is not obtained in advance, but the signal component is measured at the same time that the noise component has the frequency component in the frequency band in which the signal component does not substantially exist, and at the same time, the component ratio NCR. Seeking. Of course, the component ratio NCR measurement and the signal measurement may be performed separately in time. In the figure, BPF-1 and BPF-2 are band filters. The essential difference from the processing of FIG. 3 is that the outputs of detection means-1 (QDM-1, detection means-2 (QDM-2) are bandpass filter-1 (BPF-1), bandpass filter-2 (BPF- 2), and the frequency component of the noise component included in the measurement target physical quantity-1 (PQM-1) and the measurement target physical quantity-2 (PQM-2) is extracted.
[0051]
As described above, the component ratio NCR can be measured simultaneously with the signal measurement or at another time. Further, according to the procedure of FIG. 4, even when the component ratio NCR changes with time sufficiently more slowly than the time required for signal measurement, the average time of the average processing performed in the component ratio measurement procedure RMP is appropriately set. Therefore, it is possible to effectively and efficiently remove noise.
[0052]
(Basic items of position measurement method using magnetic field measurement)
As a reference for explaining an embodiment of the present invention, basic matters concerning a position measuring method using a magnetic field source will be described.
The method for measuring the position using a magnetic field is the same in principle, but is broadly divided into a method for obtaining the position of the magnetic field generation source and a method for obtaining the position of the magnetic field detection means for detecting the magnetic field. The
[0053]
(Method to determine the position of the magnetic field source)
In most methods for obtaining the position of the magnetic field generation source, the magnetic field generation source is single and has a small shape. In this case, in order to obtain the position and orientation of the magnetic field generation source at the same time, it is necessary to determine six unknowns, and at least six magnetic field sizes with different positions or directions are required. Of course, in this method, it is assumed that the positions and postures of the six magnetic field detecting means for detecting the magnetic field are known.
In addition, the position here is, for example, the x coordinate, the y coordinate, and the z coordinate in an appropriate coordinate system fixed to the ground. The posture is a rotation angle with respect to the x axis, the y axis, and the z axis in the same coordinate system.
[0054]
If a part of the information on the attitude of the magnetic field generation source is known by another method, the number of magnetic fields to be measured can be reduced accordingly. As a method of knowing the attitude of the magnetic field generation source, there is a method of referring to the gravity direction of the earth. In this method, since the inclination of the magnetic field generation source with respect to the vertical direction, that is, the rotation angles about two axes orthogonal to each other and orthogonal to each other can be known, two values of the three rotation angles are obtained. It is possible to know. However, the rotation angle in the vertical direction cannot be known.
If two rotation angles can be known with reference to the vertical direction, the number of measurement magnetic fields may be at least four.
[0055]
In principle, it is known what kind of spatial distribution of the magnetic field is generated by the magnetic field generation source, and if at least a part of the relationship between the relative position and the relative posture is known, a plurality of It may be a magnetic field generation source. For example, if there are two magnetic field generation sources, the information on the magnetic field obtained by one magnetic field detection means is doubled, and the unknowns regarding the magnetic field generation source are doubled to twelve. If the relationship between the relative position and the relative posture is completely unknown, 12 magnetic field magnitudes, that is, 6 magnetic field measuring means are required, which is different from measuring individual magnetic field sources. There will be no. If the relative positions of the two magnetic field generation sources are known, the number of unknowns is nine, and therefore nine magnetic fields may be measured by at least five magnetic field measuring means. Further, if the relative positions and orientations of the two magnetic field generation sources are all known, the number of unknowns is 6, so that it is only necessary to measure 6 magnetic fields with a minimum of 3 magnetic field measuring means.
[0056]
(Method for obtaining the position of the magnetic field detection means):
When obtaining the position of the magnetic field detection means, the magnetic field detection means often detects a plurality of magnetic fields at one position, such as one triaxial magnetic sensor. In this case as well, there are six unknowns: the position and attitude angle of the magnetic field detection means. Therefore, if there are two magnetic field generation sources whose relative positions and orientations are known, six magnetic fields are measured by one three-axis magnetic sensor to determine the position and posture angle of the three-axis magnetic sensor. It is possible. Here, the three-axis magnetic sensor is a sensor that can detect magnetic fields in three directions orthogonal to each other at substantially the same position in space.
[0057]
Even when measuring two of the attitude angles of the three-axis magnetic sensor with respect to the vertical direction, there are four unknowns to be obtained, but the position of the three-axis magnetic sensor is determined only from the magnetic field generated by the magnetic field generation source. In order to obtain, two magnetic field generation sources are required. However, there is a method in which only one magnetic field generation source is used and the remaining posture angle is obtained from some assumption or the direction of geomagnetism, and the position is obtained from three magnetic fields.
[0058]
On the contrary, it is also possible to use six magnetic field detecting means and one magnetic field generating source whose relative positional relationships are known. Furthermore, when measuring two posture angles of these magnetic field detection means groups with reference to the vertical direction, the number of magnetic field detection means can be four.
[0059]
In any of the above methods, the number of magnetic field generation sources and magnetic field detection means may be large. At that time, the position may be determined by a method for obtaining an optimum solution such as a least square method. These position measurement methods can use either an alternating magnetic field or a static magnetic field, but in the case of using a static magnetic field, there is a means for distinguishing the artificially generated magnetic field from the magnetic field and the geomagnetism. is necessary. Generally, this method is generally a method of measuring the magnetic field twice by reversing the direction of the artificially generated magnetic field and taking the difference between the two measurement results.
[0060]
(Example 1 regarding position measurement method)
The signal measurement method of the present invention can be applied to any of the methods using an alternating magnetic field among the position measurement methods described in the previous section (method for obtaining the position of the magnetic field detection means). That is, when an alternating magnetic field artificially generated by one or more magnetic field generation sources is detected by one or more magnetic field detection means, and at least one of the magnitude of the detected magnetic field and the position of the magnetic field detection means is changed In a position measurement method for calculating or detecting the position of the magnetic field generation source or the position of the magnetic field detection means based on the change in the magnitude, two or more magnetic field components having different ratios of the magnitude of the noise component to the signal component are detected. By using the signal measuring method of the present invention for the measured quantity detected by the magnetic field detecting means, the position measuring method having a noise removal function can be obtained.
[0061]
An example of a general configuration of a position measurement method having a noise removal function according to the present invention is shown in FIG. In the figure, MFS-1, MFS-2, MFS-3, ..., MFS-k, ..., MFS-n are magnetic field generation sources (magnetic field generation source-1, magnetic field generation source-2, magnetic field generation source-3, magnetic field , MFD-1, MFD-2, MFD-3, ..., MFD-j, ..., MFS-m are magnetic field detection means (magnetic field detection means-1, magnetic field detection). Means-2, magnetic field detection means-3, magnetic field detection means-j,..., Magnetic field detection means-m), QPS-1, QPS-2, QPS-3,..., QPS-j,. Amount (measurement amount-1, measurement amount-2, measurement amount-3, ..., measurement amount-j, ..., measurement amount-m), ADM-3, ..., ADM-j, ADM-m are posture detection means ( Attitude detection means-3, ..., Attitude detection means-j, ..., Attitude detection means-m), PLD-3, ..., PLD-j, ..., PLD-m are synchronous detectors. (Synchronous detection means -3, ..., synchronous detection means -j, ..., synchronous detection means -m), RSG reference signal generating means, PCM is the position calculating means.
[0062]
The magnetic field generation sources MFS-1, MFS-2, MFS-3,..., MFS-k,..., MFS-n are, for example, a coil, an electric wire, a looped electric wire, a mechanically rotating magnet, or the like through which an alternating current flows. Yes, an alternating magnetic field with a predetermined frequency is generated. This AC magnetic field may be subjected to amplitude modulation, phase modulation, frequency modulation, etc. for loading information. Magnetic field detection means MFD-1, MFD-2, MFD-3,..., MFD-j,..., MFD-m detect the magnetic field detected by detecting an alternating magnetic field such as an air core or a coil with a core or a fluxgate magnetometer. Any device can be used as long as it can output a signal or information proportional to the size. Attitude detection means ADM-3, ..., ADM-j, ..., ADM-m are acceleration sensors, tilt sensors, rotation sensors, gyros, etc., and magnetic field detection means MFD-3, ..., MFD-j, ..., MFD, respectively. -M is fixed at a predetermined attitude, and detects the attitude of these magnetic field detection means. The position calculation means PCM is the magnitude of the signal magnetic field obtained by the synchronous detection means PLD-3, ..., PLD-j, ..., PLD-m and the attitude detection means ADM-3, ..., ADM-j, ..., ADM-m. From the attitude angles of the magnetic field detection means MFD-1, MFD-2, MFD-3, ..., MFD-j, ..., MFD-m obtained by the above, the magnetic field generation sources MFS-1, MFS-2, MFS-3, ..., MFS-k, ..., MFS-n or magnetic field detection means MFD-1, MFD-2, MFD-3, ..., MFD-j, ..., MFD-m are calculated. NFS is a noise magnetic field generation source and is related to the object of the present invention, but is not a constituent requirement.
[0063]
The reference signal generating means RSG has a configuration example as shown in FIGS. 6 and 7, and includes two measurement amounts PQS-1 (measurement amount-1) and PQM-2 (measurement amount−) having different signal-to-noise ratios. 2) is used as an input to generate a reference signal REF that is proportional to the time variation of the signal component.
[0064]
In the configuration of FIG. 6, the reference signal generating means RSG is a portion surrounded by a broken line. Measurement target physical quantities PQM-1 (measurement target physical quantity-1) and PQM-2 (measurement target physical quantity-2), which are magnetic fields, are respectively measured by the magnetic field detection means MFD-1 and MFD-2 as measurement quantities PQS-1 and PQS-2. Converted. The changeover switch MSW switches between component ratio measurement and reference signal generation, and FIG. 6 shows a configuration in the case of component ratio measurement. When the reference signal is generated, the two switches are switched to the lower terminal, that is, to switch the measurement amount to the multiplication means.
[0065]
The component ratio measuring means RMM performs processing for obtaining the component ratio according to the equation (16) or (17), and is configured by hardware, configured by software, or a mixture of both. It doesn't matter. The obtained component ratio value is stored in hardware or software and used to generate the reference signal REF.
[0066]
The generation of the reference signal REF is obtained by multiplying the measurement amount-2 (PQS-2) by the component ratio r by the multiplication unit PRD and subtracting it from the measurement amount-1 (PQS-1) by the addition unit ADD. This is done by creating a time-varying signal in phase with the component. Further, if necessary, the waveform shaping means WFT performs processing such as rectangular wave to obtain a reference signal REF.
[0067]
In the configuration of FIG. 7, it is possible to simultaneously generate the reference signal REF and measure the component ratio r. Similar to FIG. 6, the reference signal generating means RSG is surrounded by a broken line. In order to simultaneously generate the reference signal REF and measure the component ratio r, the measurement amounts PQS-1 and PQS-2 are respectively obtained by the bandpass filters BPF-1 (bandpass filter-1) and BPF-2 (bandpass filter-2). The component of a specific frequency band is extracted. The frequency band can be determined as follows.
(1) The signal is fixed to a band that has substantially no frequency component.
(2) Either or both of the center frequency and the bandwidth can be appropriately changed according to the measurement environment, that is, the noise situation.
(3) Provide a function of dynamically changing the frequency band to maximize the noise component.
[0068]
The constituent elements of the reference signal generating means RSG in FIGS. 6 and 7 may be either software, hardware, or a mixture of both.
The synchronous detection means PLD-3,..., PLD-j,..., PLD-m are as shown in FIG. 8 as one configuration example, and the measured quantities PQS-1, PQS-2, PQS-3,. ..., PQS-m is multiplied by the reference signal REF to perform processing such as low-pass filtering and averaging, and detect a signal in phase with the reference signal REF in these measured quantities.
[0069]
FIG. 8 shows an example of the configuration of the synchronous detection means PLD. The synchronous detection means PLD is a portion surrounded by a broken line. In the figure, a measurement target physical quantity PQM, which is a magnetic field, is detected by a magnetic field detection means MFD to obtain a measurement quantity PQS. The low frequency component of the magnetic field obtained by multiplying the measured amount PQS by the reference signal REF using the multiplication means PRD is taken out by the low pass filter LPF, and averaged by the integrator ITG to obtain the signal SIG. Here, the low-pass filter LPF is not always necessary. The constituent elements of the synchronous detection means PLD in FIG. 8 may be either software, hardware, or a mixture of both.
[0070]
(Example 2 regarding position measurement method)
FIG. 9 shows a configuration of an embodiment of a position measuring apparatus having a noise removal function according to the present invention, which measures the position of a magnetic field generation source.
MFS is a magnetic field generation source, ADM is posture detecting means, and FRM is a frame. In this embodiment, magnetic field detection means MFD-1, MFD-2, MFD-3, attitude detection means ADM, reference signal generation means RSG, synchronous detection means PLD-1, PLD-2, PLD-3, position calculation means PCM Are fixed to the frame FRM. The position of the magnetic field generation source MFS is measured by moving the frame FRM so that the magnetic field intensity generated in the specific direction is maximized in a place where the magnetic field generated by the magnetic field generation source MFS satisfies a specific condition, for example, in a certain horizontal plane. A place is searched, and the position of the magnetic field generation source MFS is referred to or estimated from the position and the magnetic field intensity there.
[0071]
Since the magnetic field detection means-1 (MFD-1) and the magnetic field detection means-2 (MFD-2) are used both for generating a reference signal and for position measurement, the reference signal generation means is as shown in FIG. It must be configured.
The posture detection means ADM does not necessarily have to be provided as hardware. A measurer moving the frame FRM may perform an equivalent function by maintaining the posture of the frame RFM in a predetermined direction. Further, at least two magnetic field detecting means MFD are sufficient. In this case, the magnetic field detection means-3 (MFD-3) is excluded, and there is an advantage that the number of components is small.
[0072]
(Example 3 regarding position measurement method)
FIG. 10 shows an embodiment of the present invention in which magnetic field detection means are installed at a plurality of spatially separated positions to measure the position of the magnetic field generation source MFS. FIG. 10 shows a configuration in the case where magnetic field detection means are placed at three spatially separated locations.
In the figure, UMM-1, UMM-2 and UMM-3 are unit measuring means used as magnetic field detecting means. The PMM is position measuring means, and is composed of signal receiving means SRX and position calculating means PCM. In this embodiment, in order to place the magnetic field detecting means at a plurality of spatially separated positions, unit measuring means UMM-1 (unit measuring means-1) and UMM-2 (unit measuring means-2) having a signal transmission function. UMM-3 (unit measuring means-3) is used. As a result, signals are exchanged with the signal receiving means SRX in the position measuring means PMM.
[0073]
A configuration example of the unit measuring means UMM is shown in FIGS. In the example of FIG. 11, the unit measuring means UMM includes magnetic field detecting means MFD-1, MFD-2, MFD-3, attitude detecting means ADM, reference signal generating means RSG, synchronous detecting means PLD-1, PLD-2, PLD-. 3. It consists of signal transmission means STX. The magnetic field detection means MFD-1, MFD-2, and MFD-3 are not necessarily arranged so as to detect magnetic fields in directions orthogonal to each other as long as they can detect independent magnetic fields. As in the present invention, the magnetic field detection means MFD-1, MFD-2, and MFD-3 constitute a three-axis magnetic sensor that detects magnetic field strengths in three different directions orthogonal to each other at substantially the same point. The attitude detection means ADM is fixed to the magnetic field detection means MFD-1, MFD-2, and MFD-3, and detects these attitude angles. The signal transmission unit STX has a function of transmitting the outputs of the synchronous detection units PLD-1, PLD-2, and PLD-3 and the output of the attitude detection unit ADM as electromagnetic field signals, voltage signals, current signals, or optical signals.
[0074]
FIG. 12 shows unit measuring means UMM for detecting magnetic fields in two orthogonal directions. Therefore, there is no magnetic field detection means MFD-3 and synchronous detection means PLD-3 compared to the configuration of FIG.
[0075]
The signal receiving means SRX in FIG. 10 is the output of the synchronous detection means PLD-1, PLD-2, PLD-3 sent from the signal sending means STX in the unit measuring means UMM (in the case of the configuration in FIG. 11) or synchronous. The outputs of the detection means PLD-1 and PLD-2 (in the case of the configuration of FIG. 12) and the output of the attitude detection means ADM are received and passed to the position calculation means PCM.
[0076]
FIG. 13 shows an embodiment in which there are two unit measuring means, unit measuring means-1 (UMM-1) and unit measuring means-2 (UMM-2).
[0077]
When independent six magnetic field magnitudes are required, the unit measuring means UMM having the configuration shown in FIG. 12 is used in the embodiment of FIG. 10, or the unit measuring means having the configuration shown in FIG. 11 is used in the embodiment of FIG. You may use UMM. Of course, there is no problem even if the unit measuring means UMM having the configuration of FIG. 11 is used in the configuration of FIG.
[0078]
Further, in the embodiment of FIGS. 10 and 13, the unit measuring means UMM having the configuration of FIG. 14 can be used in place of the unit measuring means UMM having the configuration of FIG.
[0079]
In the unit measuring means UMM of FIG. 14, the two magnetic field detecting means, the magnetic field detecting means-1 (MFD-1) and the time detecting means-2 (MFD-2) are placed in spatially different places. Attitude detection means for detecting the attitude of these magnetic field detection means, attitude detection means-1 (ADM-1), and attitude detection means-2 (ADM-2) are respectively magnetic field detection means-1 (MFD-1). , Magnetic field detection means-2 (MFD-2). Output signal of magnetic field detection means-1 (MFD-1) and output signal of attitude detection means-1 (ADM-1), and output signal of magnetic field detection means-2 (MFD-2) and attitude detection means-2 ( The output signals of ADM-2) are respectively input to the detection signal processing means SPM and subjected to the same processing as in the case of FIG. 11 and FIG.
[0080]
(Example 4 regarding position measurement method)
FIG. 15 shows an embodiment in which four independent magnetic field magnitudes are required because information on two attitude angles of the magnetic field generation source can be used. In this case, the unit measuring means UMM having the configuration shown in FIG. 12 may be used. Of course, there is no problem even if the unit measuring means UMM configured as shown in FIG. 11 is used.
[0081]
(Example 5 regarding position measurement method)
FIG. 16 shows a configuration for carrying out a measurement method when measuring the position of the magnetic field detection means using two magnetic field generation sources whose mutual positions and orientations are known.
Magnetic field source-1 (MFS-1) and magnetic field source-2 (MFS-2)
(1) Generate magnetic fields with different frequencies,
(2) generate magnetic fields coded with different signs;
(3) generate a magnetic field at different times;
The AC magnetic field is generated in such a way that the magnetic field generated by each magnetic field generation source can be identified by a method such as
The magnetic field detection means MFD-1, MFD-2, and MFD-3 constitute a three-axis magnetic sensor that detects magnetic field strengths in three different directions orthogonal to each other at substantially the same point.
[0082]
(Example 6 regarding position measurement method)
FIG. 17 shows another configuration of the measurement method in the case of measuring the position of the magnetic field detection means using two magnetic field generation sources whose mutual positions and orientations are known. In this embodiment, the part for measuring the magnetic field strength and the part for calculating the position are placed at spatially separated locations.
[0083]
【The invention's effect】
As described above in detail, according to the present invention, it can be widely applied to the removal of noise when measuring a physical quantity through a magnetic field, an electric field, etc., and affects the measurement like an underground power line. Even when the noise current to be applied is close, it is possible to measure the position with high accuracy, and it is possible to measure the excavation position when performing horizontal drilling etc. in urban areas where there is a lot of magnetic noise. The effect is extremely large.
[Brief description of the drawings]
FIG. 1 is a block diagram for explaining a processing procedure for obtaining a component ratio when there is a signal vector and a noise vector used in the method of the present invention.
FIG. 2 is a block diagram for explaining a processing procedure for obtaining a component ratio when there is a signal vector and a noise vector used in the method of the present invention.
FIG. 3 is a block diagram for explaining a procedure for obtaining a component ratio when only noise exists in the method of the present invention;
FIG. 4 is a block diagram for explaining a procedure for obtaining a component ratio by detecting noise in a frequency band in which no signal component exists in the method of the present invention.
FIG. 5 is a block diagram illustrating a configuration example for carrying out a position measurement method having a noise removal function according to the present invention.
FIG. 6 is a block diagram showing an example of reference signal generating means used in the present invention.
FIG. 7 is a block diagram showing another example of reference signal generating means used in the present invention.
FIG. 8 is a block diagram showing a configuration example of synchronous detection means located in the algae in the present invention.
FIG. 9 is a block diagram showing another configuration example for implementing the position measurement method having a noise removal function according to the present invention.
FIG. 10 is a block diagram showing another configuration example for implementing the position measurement method having a noise removal function according to the present invention.
11 is a block diagram showing one configuration example of unit measuring means used in the position measuring method having a noise removal function according to the present invention shown in FIG.
12 is a block diagram showing another configuration example of the unit measuring means used in the position measuring method having a noise removal function according to the present invention shown in FIG.
FIG. 13 is a block diagram showing still another configuration example for implementing the position measurement method having a noise removal function according to the present invention.
14 is a block diagram showing a configuration example of unit measurement means used in the configuration example of FIG. 10 or FIG. 13;
FIG. 15 is a block diagram showing still another configuration example for implementing the position measurement method having a noise removal function according to the present invention.
FIG. 16 is a block diagram showing a configuration example for measuring the position of the magnetic field detection means according to the present invention.
FIG. 17 is a block diagram showing another configuration example for measuring the position of the magnetic field detection means according to the present invention.
FIG. 18 is a schematic diagram for explaining a noise removal method in conventional biomagnetism measurement.
[Explanation of symbols]
PQM-1 to PQM-M Physical quantities to be measured 1 to M
QDM-1 to QDM-M detection means 1 to M
RMP component ratio measurement procedure
NCR component ratio
PRD-0 to PRD-M multiplication means 0 to M
ADD addition means
WET waveform shaping means
REF reference signal
AVR-3 to AVR-M Averaging means-3 to M
SIG-3 to SIG-M Signal-3 to M
BPF-1, BPF-2 Bandpass filter 1, 2
MFS-1 to MFS-n Magnetic field source 1 to n
MFD-1 to MFD-m Magnetic field detection means 1 to m
PQS-1 to PQS-m Measurement amount-1 to m
RSG reference signal generating means
PLD-1 to PLD-m Synchronous detection means-1 to m
PCM position calculation means
RMM component ratio measuring means
LPF low-pass filter
ITG integrator
FRM frame
NFS Noise magnetic field source
SRX signal receiving means
PMM signal measuring means
ADM attitude detection means
UMM-1 to UMM-3 unit measuring means -1 to 3
UMP-1 to UMP-2 unit measuring means pair 1-2
SPM detection signal processing means
STX signal sending means

Claims (20)

Each of at least two measurement target physical quantities in which a signal component whose magnitude is to be measured and a noise component are linearly superimposed is detected by a detection unit that senses the measurement target physical quantity and outputs a signal proportional to the measurement target physical quantity. The signal components of the first measurement quantity and the second measurement quantity among the measurement quantities obtained by detection change with the same time, and the respective noises of the first measurement quantity and the second measurement quantity are detected. When the components change the same time,
A value obtained by multiplying the second measured quantity by a noise component ratio, which is a ratio of the magnitude of the noise component in the first measured quantity to the magnitude of the noise component in the second measured quantity. As a reference signal, subtract from the measured amount of
By synchronous detection of the measurement amount of the measurement target physical quantity of interest, a signal measurement how to measure the magnitude of the signal component in the measured physical quantity,
The measurement target physical quantity is one or more components of a vector quantity, and the measurement target physical quantity is constituted by a sum of a signal vector and a noise vector;
The signal vector is represented as a unit quantity giving the direction of the signal vector as a scalar quantity multiple giving the time change of the signal vector;
A unit vector giving the direction of the noise vector is represented as a scalar quantity multiple giving a time variation of the noise vector which is not completely identical to the signal vector, and
When the direction of the signal vector and the direction of the noise vector are not parallel,
Two spatially orthogonal two of the measurement target physical quantities that are not necessarily measured at the same spatial position by the detection means that senses the measurement target physical quantity and outputs a signal proportional to the size of the component. By obtaining a measurement amount proportional to the magnitude of the above component, in the first course, an amount obtained by dividing the magnitude of the second component by the magnitude of the first component of the unit vector that gives the direction of the noise vector By multiplying the noise component ratio by the measurement amount related to the first component of the measurement target physical quantity and subtracting it from the measurement amount related to the second component of the measurement target physical quantity, the signal vector is proportional to the scalar quantity changing with time. Get the reference signal
In the second process, the signal component in the detection target component of the measurement target physical quantity is averaged based on the amplitude of the reference signal by averaging the reference signal multiplied by the measurement quantity that is the detection target of the measurement target physical quantity. A signal measuring method characterized by knowing the size of the signal.   The signal measurement method according to claim 1, wherein the measurement quantity is obtained by detecting the physical quantity to be measured at two or more different locations. Signal measurement method according to claim 1 or 2, characterized in that the scalar amount giving a time variation of the signal vector is periodic signal. 3. The signal measuring method according to claim 1, wherein the signal component is a signal subjected to amplitude modulation, frequency modulation, or phase modulation. When the average level of the reference signal is a threshold value, a positive value of a predetermined magnitude is given when the signal level of the reference signal is higher than the threshold value, and a negative value of the same magnitude is given when lower than the threshold value, Or depending on whether the signal is fully amplified and truncated at a given positive value and negative value of the same magnitude,
The rectangular wave reference signal is obtained by averaging the rectangular wave reference signal obtained by multiplying the rectangular wave reference signal as a rectangular wave reference signal by multiplying the rectangular wave reference signal by multiplying the square wave reference signal by the measurement quantity that is the detection target of the measurement target physical quantity. signal measurement method according to claim 1 or 2 based on the amplitude, characterized in knowing the magnitude of the signal component in the detection target component in the measurement target physical quantity. The time variation of the signal vector of the reference signal is periodic, the rectangular wave reference signal is divided into predetermined time intervals, and the period of the reference signal is constant within each time interval. The signal measurement method according to claim 5 , wherein the period of the reference signal is adjusted as described above. The time variation of the signal vector of the reference signal is a period, and the period of the reference signal is adjusted so that each period of the rectangular wave reference signal is an average of a predetermined number of periods including the same period. The signal measurement method according to claim 5 . Signal measurement method according to any of claims 1 to 5, characterized in that previously obtained from the measurement target physical quantity when the signal component is not a The noise component ratio. When the scalar component giving the temporal change of the noise component has a frequency component in the frequency domain where the frequency component of the scalar component giving the temporal change of the signal component has a sufficiently low level, the physical quantity to be measured in the frequency domain signal measurement method according to any of claims 1 to 7, wherein the determination of the the noise component ratio by component ratio. The signal measurement method according to claim 9 , wherein the frequency band is fixed in advance. The signal measurement method according to claim 9 , wherein the frequency band is variable. The signal measurement method according to claim 9, further comprising a function of dynamically selecting the frequency band so that a noise component is maximized. Signal measurement method according to any one of claims 1 to 12, the measured physical quantity characterized in that it is a magnetic field or magnetic flux density. An AC magnetic field artificially generated by at least one magnetic field generation source is detected by at least one magnetic field detection means, and the magnitude when the detected magnetic field magnitude and / or the position of the magnetic field detection means is changed. For the position measurement to calculate or detect the position of the magnetic field generation source or the position of the magnetic field detection means from the change in length,
The magnetic field detection means is arranged so as to detect a plurality of magnetic field components having different ratios of noise components to signal components, and the measured quantity detected by the magnetic field detection means is any one of claims 1 to 14. a position measuring method allowed have a noise removal function by applying a signal measurement method described,
The magnetic field source is one;
At least four magnetic field detecting means are arranged at least at three different points at least at two different points, and the directions of magnetic fields detected by the magnetic field detecting means arranged at each point are vector-independent from each other. Arranged to be
The magnetic field generated by the magnetic field generation source is detected by the magnetic field detection means, and the position of the magnetic field generation source is determined from the attitude of the magnetic field detection means, the magnitude of the magnetic field detected by the magnetic field detection means, and the attitude of the magnetic field generation source. A position measurement method having a noise removal function, characterized in that The number of magnetic field generation sources is one, the plurality of magnetic field detection means can be moved without changing the relative position and posture of each magnetic field detection means, and magnetic fields in a plurality of spatially orthogonal directions are used. Fixed to the frame so that it can be detected,
Detecting the position of the magnetic field generation source by detecting the magnitude of the alternating magnetic field generated by the magnetic field generation source and the change in the magnitude by the magnetic field detection means fixed to the frame while moving the frame. The position measurement method having a noise removal function according to claim 14 . The magnetic field source is one;
At least six magnetic field detecting means are arranged at least at two different points, and at least six magnetic field detecting means are arranged, and the directions of the magnetic fields detected by the magnetic field detecting means arranged at each point are vector-independent from each other. Arranged to be
The magnetic field generated by the magnetic field generation source is detected by the magnetic field detection means, and the position and posture of the magnetic field generation source are calculated from the attitude of the magnetic field detection means and the magnitude of the magnetic field detected by the magnetic field detection means. The position measurement method having a noise removal function according to claim 14 . 15. The noise removal according to claim 14 , wherein the magnetic field generation source is installed in or near the excavation head used in the non-cutting method, and the position of the magnetic field generation source is measured. A position measuring method having a function. There are two or more magnetic field generation sources in which at least one of the mutual positional relationship and posture relationship is known;
The magnetic field generated by the magnetic field generation source is detected by at least three magnetic field detection means for detecting a magnetic field in three directions that are linearly independent in vector, and the attitude of the magnetic field detection means and the magnetic field detection means are used. 15. The position measurement method having a noise removal function according to claim 14 , wherein the position of the magnetic field detection means is calculated from the detected magnitude of the magnetic field and the position of the magnetic field generation source. By generating an alternating magnetic field at different times or by generating alternating magnetic fields with different frequencies, it is possible to generate alternating magnetic fields that are distinguishable from each other, and at least one of the mutual positional relationship or posture relationship is known There are two or more magnetic field generation sources, and each magnetic field generation source generates an alternating magnetic field in such a manner that the alternating magnetic field generated by each magnetic field generation source can be individually identified
The magnitude of the magnetic field detected by the magnetic field detection means by detecting the magnetic field generated by each magnetic field generation source with at least three magnetic field detection means for detecting magnetic fields in three directions that are linearly independent in vector. 15. The position measurement method having a noise removal function according to claim 14 , wherein the position of the magnetic field detection means is detected from the position of the magnetic field generation source. Are those wherein the magnetic field detecting means is provided in the vicinity of the drilling head or drilling head used in trenchless method of claim 18 or 19, characterized in that to measure the position of the magnetic field generating source A position measurement method having a noise removal function.

Images