JP3612735B2 - Vehicle noise reduction device and control signal setting method - Google Patents

Description

（式（１）中、Ｗｐ＝１／ｂ、‖Ｑ‖ _∞ は、クローズドループにおける目標値から制御量までの伝達関数Ｑの最大のゲインであってｓｕｐ _W ‖Ｑ（ｊｗ）‖により定義されるものであり、‖・‖は行列のノルムを表す。）
【００１３】
請求項２の発明では、上記請求項１の発明において、制御手段は、騒音検出手段の出力信号に代えて制御信号からオーディオ装置のオーディオ信号を除去するように構成されているものとする。
【００１４】
請求項３の発明では、上記請求項１の発明において、制御手段は、騒音検出手段の出力信号に加えて制御信号からもオーディオ装置のオーディオ信号を除去可能に構成されているものとする。そして、上記オーディオ装置のオーディオ信号を騒音検出手段の出力信号又は制御手段の制御信号から選択的に除去するために切り換える切換手段を設ける。
【００１５】
請求項４の発明では、上記請求項１の発明と同じ前提に立ちかつオーディオ装置のオーディオ信号が制御手段に入力されているものとした上で、上記制御手段は、上記低減音発生手段及び騒音検出手段間の音の伝達特性に関して予め行われた測定結果に基づいてモデル化してなる制御音伝達特性モデルＨと、上記測定結果と上記制御音伝達特性モデルとの差の絶対値に基づいて予め定められたモデル化誤差値Ｗｕと、上記騒音伝達特性Ｇと、該騒音伝達特性の低減騒音レベルを示すために予め定められた騒音低減用しきい値ｂと、定数Ｋとが、下記式（１）を満たすように定数Ｋを求めて、該定数Ｋと上記騒音検出手段の出力信号とに基づいて上記制御信号を設定し、かつ上記オーディオ装置の作動時には、上記制御信号の設定を中止するように構成されているものとする。
【００１６】
【数６】

（式（１）中、Ｗｐ＝１／ｂ、‖Ｑ‖ _∞ は、クローズドループにおける目標値から制御量までの伝達関数Ｑの最大のゲインであってｓｕｐ _W ‖Ｑ（ｊｗ）‖により定義されるも のであり、‖・‖は行列のノルムを表す。）
【００１７】
請求項５の発明では、オーディオ装置を有する車室内に設置位置における騒音源からの騒音を検出する騒音検出手段を設置し、この騒音検出手段の出力信号に基づき、騒音源及び騒音検出手段間の騒音伝達特性を低下させて乗員に対する騒音を低減するための制御信号を設定し、この制御信号を低減音発生手段に出力して騒音低減音を発生させるようにした車両騒音低減装置における上記制御信号の設定方法として、上記低減音発生手段及び騒音検出手段間の音の伝達特性に関して予め行われた測定結果に基づいてモデル化してなる制御音伝達特性モデルを定め、上記測定結果と上記制御音伝達特性モデルとの差の絶対値に基づいてモデル化誤差値を定め、上記騒音伝達特性の低減騒音レベルを示すための騒音低減用しきい値を定めておき、上記制御音伝達特性モデルＨと上記モデル化誤差値Ｗｕと上記騒音伝達特性Ｇと上記騒音低減用しきい値ｂと定数Ｋとが、下記式（１）を満たすように定数Ｋを求めて、該定数Ｋと上記騒音検出手段の出力信号とに基づいて上記制御信号を設定するとともに、上記オーディオ装置の作動時に、騒音検出手段の出力信号からオーディオ装置のオーディオ信号を除去することとする。
【００１８】
【数７】

（式（１）中、Ｗｐ＝１／ｂ、‖Ｑ‖ _∞ は、クローズドループにおける目標値から制御量までの伝達関数Ｑの最大のゲインであってｓｕｐ _W ‖Ｑ（ｊｗ）‖により定義されるものであり、‖・‖は行列のノルムを表す。）
【００１９】
請求項６の発明では、上記請求項５の発明において、騒音検出手段の出力信号に代えて、制御信号からオーディオ装置のオーディオ信号を除去するようにする。
【００２０】
請求項７の発明では、上記請求項５の発明において、騒音検出手段の出力信号に加えて、制御信号からもオーディオ装置のオーディオ信号を除去するようにし、上記オーディオ装置のオーディオ信号を騒音検出手段の出力信号又は制御信号から選択的に除去するために切り換えるようにする。
【００２１】
請求項８の発明では、上記請求項５の発明と同じ前提に立ち、上記低減音発生手段及び騒音検出手段間の音の伝達特性に関して予め行われた測定結果に基づいてモデル化してなる制御音伝達特性モデルを定め、上記測定結果と上記制御音伝達特性モデルとの差の絶対値に基づいてモデル化誤差値を定め、上記騒音伝達特性の低減騒音レベルを示すための騒音低減用しきい値を定めておき、上記制御音伝達特性モデルＨと上記モデル化誤差値Ｗｕと上記騒音伝達特性Ｇと上記騒音低減用しきい値ｂと定数Ｋとが、下記式（１）を満たすように定数Ｋを求めて、該定数Ｋと上記騒音検出手段の出力信号とに基づいて上記制御信号を設定するとともに、上記オーディオ装置の作動時には上記制御信号の設定を中止するようにする。
【００２２】
【数８】

（式（１）中、Ｗｐ＝１／ｂ、‖Ｑ‖ _∞ は、クローズドループにおける目標値から制御量までの伝達関数Ｑの最大のゲインであってｓｕｐ _W ‖Ｑ（ｊｗ）‖により定義されるものであり、‖・‖は行列のノルムを表す。）
【００２３】
【作用】
請求項１又は５の発明では、車室内において、騒音検出手段により騒音源からの騒音が検出される。そして、上記騒音検出手段の出力信号を受けた制御手段により制御信号が設定され、この制御信号を受けた低減音発生手段により騒音低減音が発生される。この騒音低減音により、上記騒音源及び騒音検出手段間の騒音伝達特性が低下され、このことで、乗員への騒音が低減される。このとき、上記制御手段では、制御音伝達特性モデルＨとモデル化誤差値Ｗｕと騒音伝達特性Ｇと騒音低減用しきい値ｂと定数Ｋとが、式（１）を満たすように定数Ｋが求められ、この定数Ｋと騒音検出手段の出力信号とに基づいて制御信号が設定される。すなわち、上記制御音伝達特性モデルが用いられることにより、騒音検出手段の設置位置における騒音低減音の騒音に対する低減効果が高められる。このとき、上記モデル化誤差値及び騒音低減用しきい値が考慮されることにより、騒音発生手段の出力信号が小さくなるように制御信号をフィードバック制御する際の外乱に対する安定性が確保される。
【００２４】
また、オーディオ装置の作動時には、上記騒音検出手段の出力信号からオーディオ装置のオーディオ信号が制御手段により除去され、このことで、騒音検出手段により検出される音のうち、低減対象である騒音からオーディオ音が除外される。この場合には、制御信号の設定の前にオーディオ装置本体の出力信号が除去されることにより、騒音検出手段がオーディオ音を検出していない状態にあると見做されるので、上記低減音発生手段の騒音低減音がオーディオ音に影響を及ぼすという事態は、騒音検出手段の設置位置において回避される。
【００２５】
請求項２又は６の発明では、上記騒音検出手段の出力信号に代えて、制御信号からオーディオ装置のオーディオ信号が制御手段により除去される。この場合には、制御信号の設定の後にオーディオ信号が除去されることにより、低減音発生手段の騒音低減音からオーディオ音に関係する音が除外されるので、この騒音低減音がオーディオ音に影響を及ぼすという事態は、車室内全体において回避される。
【００２６】
請求項３又は７の発明では、上記オーディオ装置のオーディオ信号は、騒音検出手段の出力信号からでも制御信号からでも切り換えて除去されるので、騒音検出手段の設置位置に限って騒音低減音のオーディオ音への影響が回避される作用と、車室内全体において騒音低減音のオーディオ音への影響が回避される作用とを、乗員が任意に切り換えることができる。
【００２７】
請求項４又は８の発明では、上記オーディオ装置の作動時には、制御信号の設定が停止される。よって、オーディオ装置の非作動時にはロードノイズ等の騒音を効果的に低減する一方、オーディオ装置の作動時には、騒音低減音がオーディオ音に悪影響を及ぼすという事態は完全に回避される。
【００２８】
【実施例】
以下、本発明の実施例を図面に基づいて説明する。
【００２９】
（実施例１）
図２及び図３は、本発明の実施例１に係る騒音低減装置の全体構成を概略的に示し、この騒音低減装置は、オーディオ装置が搭載された自動車（車両）における各座席の乗員に対するロードノイズ等、数十〜数百Ｈｚの騒音を騒音源信号（リファレンス信号）を用いずに低減するためのものである。
【００３０】
同各図において、１は自動車の車体、２は車体１の前後中央部に位置する車室、３，４は車室２を開閉する前後のドアであって、上記車室２内には前後に座席５，６が設置され、前部座席５（図２及び図３のそれぞれ左側の座席）は右側の運転席５ａ及び左側の助手席５ｂで、また後部座席６（図２及び図３のそれぞれ右側の座席）は左右２つの後席６ａ，６ａでそれぞれ構成されている。７は、運転席５ａの前方位置に設置されたステアリングホイールである。
【００３１】
そして、本実施例１では、運転席５ａ，助手席５ｂ及び後席６ａにおける各ヘッドレスト８の左右何れかの側（着座した乗員の左右何れかの耳に近い位置）がそれぞれ騒音低減箇所とされ、これらの箇所にそれぞれ騒音検出手段としてのマイクロフォン１０が１個ずつ合計４個取り付けられ、これらのマイクロフォン１０，１０，…により車室２内の音を乗員の耳近くで検出するようにしている。一方、上記車体１左右の前側ドア３，３の車室２内側面、及び車室２後端のパッケージトレイ１１の左右両側部にはそれぞれ車室２内に騒音低減音を発生させる低減音発生手段としてのスピーカ１２，１２，…が配置され、これら４個のスピーカ１２，１２，…はオーディオ装置１５と兼用されている。そして、インストルメントパネル１３の左右中央に、オーディオ装置１５が設置されている。
【００３２】
上記スピーカ１２，１２，…及びマイクロフォン１０，１０，…は、例えば助手席５ｂ側のインストルメントパネル１３内に配置した制御手段としてのコントローラ１４に接続されている。そして、上記車室２内の騒音と各スピーカ１２から発せられる騒音低減音との合成音を各マイクロフォン１０で検出し、そのマイクロフォン１０から出力されるマイク信号ｕをコントローラ１４に入力するとともに、各スピーカ１２にマイク信号ｕとは逆位相である制御信号としてのスピーカ信号ｙを出力することにより、各マイクロフォン１０の位置で各スピーカ１２からの騒音低減音を騒音と干渉させて、各マイクロフォン１０により検出される音を低減するようにしている。
【００３３】
上記コントローラ１４は、図４に詳しく示すように、デジタル信号処理によりマイク信号ｕを小さくするためのスピーカ信号ｙを出力する制御手段としてのＣＰＵ１６（セントラル・プロセシング・ユニット）を有し、該ＣＰＵ１６の入力段には、各マイクロフォン１０からのマイク信号ｕを増幅するマイクアンプ１７と、増幅されたマイク信号ｕの低周波部分（例えば５００〜１０００Ｈｚ以下）を瀘波するローパスフィルタ１８と、瀘波されたアナログのマイク信号ｕをデジタル信号に変換するＡ／Ｄ変換部１９とがそれぞれマイクロフォン１０，１０，…の個数と同じ数だけ順に接続されている。一方、ＣＰＵ１６の出力段には、デジタルのスピーカ信号ｙをアナログ信号に変換するＤ／Ａ変換部２０と、アナログ信号に変換されたスピーカ信号ｙの低周波部分を瀘波するローパスフィルタ２１と、瀘波されたスピーカ信号ｙを増幅するスピーカアンプ２２とがそれぞれスピーカ１２，１２，…の個数と同じ数だけ順に接続されている。また、上記ＣＰＵ１６の入力段には、オーディオ装置本体１５からのオーディオ信号ｏを増幅するマイクアンプ２３と、増幅されたオーディオ信号ｏの低周波部分（例えば５００〜１０００Ｈｚ以下）を瀘波するローパスフィルタ２４と、瀘波されたアナログのオーディオ信号ｏをデジタル信号に変換するＡ／Ｄ変換部２５とが順に接続されている。上記演算処理ブロック１６、Ａ／Ｄ変換部１９，２５及びＤ／Ａ変換部２０の各作動は、図外のサンプリングクロック発生部で発生したサンプリング周期信号により互いに同期して行われる。
【００３４】
上記ＣＰＵ１６の構成を図５に基づいてさらに詳しく説明する。マイク信号ｕはｍ個（ここではｍ＝４）のマイク信号ｕ１〜ｕｍからなる列ベクトルであり、基本的に、これらマイク信号ｕ１〜ｕｍに基づいて、該ＣＰＵ１６に設けられた演算ブロック４０の持つ定数Ｋによりｌ個（ここではｌ＝４）のスピーカ信号ｙ１〜ｙｌからなる列ベクトルとしてのスピーカ信号ｙが演算される。上記定数Ｋは、図６に示すように、Ａ〜Ｄの４つのマトリクス成分からなる行列であり、その演算処理は、図７に示すフローチャートのように行われる。すなわち、ステップＳ１でマイク信号ｕを入力した後、ステップＳ２では、前回のマイク信号ｘにＣ成分が乗算された値と、今回のマイク信号ｕにＤ成分が乗算された値とを加算して、スピーカ信号ｙを求める。そして、ステップＳ３で上記スピーカ信号ｙをスピーカ１２に出力し、次いでステップＳ４に移る。このステップＳ４では、前回のマイク信号ｘにＡ成分が乗算された値と、今回のマイク信号ｕにＢ成分が乗算された値とを加算して、上記ステップＳ２で処理する際に前回のマイク信号ｘとなる値を求め、その後、上記ステップＳ１に戻る。
【００３５】
本発明の特徴として、上記ＣＰＵ１６は、図１（ａ）に示すように、スピーカ１２からマイクロフォン１０に達する騒音低減音の伝達特性をモデル化してなる制御音伝達特性モデルＨについて定められたモデル化誤差値Ｗｕ（Uncertainty Weight）と、図１（ｂ）に示すように、騒音源Ｓから上記マイクロフォン１０に入ってくる騒音伝達特性Ｇの低減すべき騒音レベルを示すために定められた騒音低減用しきい値ｂ（Performance Objective）とにより決定される定数Ｋに基づいてスピーカ信号ｙを設定するようになされている（尚、説明の簡単化のためにスピーカ１２，１２，…及びマイクロフォン１０，１０，…はそれぞれ１個とし、かつオーディオ装置１５及び騒音低減装置は、各々、専用のスピーカを有するものとする）。また、上記オーディオ装置１５の作動時には、マイク信号ｕからオーディオ装置１５のオーディオ信号ｏを除去するように構成されている。
【００３６】
具体的には、上記制御音伝達特性モデルＨは、ＣＰＵ１６からスピーカ信号ｙを出力した後に該スピーカ信号ｙによりスピーカ１２，１２，…がそれぞれ駆動制御されて車室２内の音に変化があり、この音の変化がマイクロフォン１０により検出されてそのマイク信号ｕがＣＰＵ１６に入力されるまでの音の伝達特性を実際の測定結果Ｈ′に基づいてモデル化したものである。そして、その際のモデル化誤差を考慮して、上記演算ブロック４０の定数Ｋを決定する際に誤差量の絶対値｜Ｈ−Ｈ′｜に対するモデル化誤差値Ｗｕを定めている。一方、上記騒音伝達特性Ｇに対し騒音低減用しきい値ｂを定めておいて、コントローラ１４を含むクローズドループの伝達関数のピークを集中的にラインｂ以下に落とすようになされている。つまり、騒音低減用重みＷｐ（Performance Weight）を、Ｗｐ＝１／ｂとすると、全ての周波数に対して、
【数９】

つまり、
【数１０】

の関係式が成り立つようにする。そして、これらモデル化誤差値Ｗｕ及び騒音低減用しきい値ｂを定めることが、上記定数Ｋを決定してスピーカ信号ｙを設定することになる。
【００３７】
このとき、上記モデル化誤差値Ｗｕと騒音低減用しきい値ｂとは互いに独立には決められない。例えば、モデル化誤差値Ｗｕを小さくして騒音低減効果を高めようとすると車室２内の音響特性の変化を受け易くなって騒音低減制御が不安定になることから、モデル化誤差値Ｗｕを大きくすると、今度は、騒音低減用しきい値ｂを下げることができなくなる。そこで、ロバスト制御の１つであるＨ∞制御理論においてフィードバック特性の場合の混合感度問題として知られている次式（１）を満たすようなモデル化誤差値Ｗｕ及び騒音低減用しきい値ｂを選択して定数Ｋを決定していく。
【００３８】
【数１１】

（式（１）中、Ｗｐ＝１／ｂ、‖Ｑ‖ _∞ は、クローズドループにおける目標値から制御量までの伝達関数Ｑの最大のゲインであってｓｕｐ _W ‖Ｑ（ｊｗ）‖により定義されるものであり、‖・‖は行列のノルムを表す。）
【００３９】
次に、上記オーディオ装置１５のオーディオ信号ｏをマイク信号ｕから除去してスピーカ信号ｙを設定する回路について図５及び図８に基づいてさらに詳しく説明する。
【００４０】
上記ＣＰＵ１６における演算ブロック４０の入力段には、図８に原理的に示すように、スピーカ１２とマイクロフォン１０との間の音の伝達特性をモデル化した制御音伝達特性モデルＨ１に基づいて、各々のスピーカ１２に出力されるオーディオ信号ｏの位相及びゲインを調整するモデルフィルタ２６と、このモデルフィルタ２６から出力されたオーディオ信号ｏをマイク信号ｕから除去する減算器２７とが設けられている。具体的には、図５に示すように、上記モデルフィルタ２６はマイク信号ｕの数（ｍ個）だけ配置され、かつ各モデルフィルタ２６はスピーカ信号ｙと同じ数（ｌ個）だけのモデルフィルタ２６ａ，２６ａ，…が並列に接続されてなっている。これら各モデルフィルタ２６ａは、各々、スピーカ１２の騒音低減音がマイクロフォン１０に達するまでの間の音の伝達特性がモデル化された制御音伝達特性モデルＨ１〜Ｈｌを備えている。つまり、第１のマイクロフォン１０については、該マイクロフォン１０とｌ個のスピーカ１２，１２，…との間の各制御音伝達特性モデルＨ１１，Ｈ２１，…，Ｈｌ１からなるモデルフィルタ２６ａ，２６ａ，…を備えたモデルフィルタ２６が、したがって、第ｍのマイクロフォン１０については、該マイクロフォン１０とｌ個のスピーカ１２，１２，…との間の各制御音伝達特性モデルＨ１ｍ，Ｈ２ｍ，…，Ｈｌｍからなるモデルフィルタ２６ａ，２６ａ，…を備えたモデルフィルタ２６がそれぞれ連設されている。また、各モデルフィルタ２６には、モデルフィルタ２６ａ，２６ａ，…の各出力値を合算してその合算値を上記減算器２７に出力する加算器２６ｂが設けられている。
【００４１】
したがって、本実施例１によれば、車室２内の運転席５ａ、助手席５ｂ及び後席６ａにおいて、マイクロフォン１０により騒音源Ｓからの騒音が検出される。そして、マイク信号ｕを受けたＣＰＵ１６によりスピーカ信号ｙが設定され、このスピーカ信号ｙを受けたスピーカ１２により騒音低減音が発生される。この騒音低減音により、上記騒音源Ｓ及びマイクロフォン１０間の騒音伝達特性Ｇが低下され、このことで、乗員への騒音ｄが低減される。このとき、上記ＣＰＵ１６において制御音伝達特性モデルＨが用いられることにより、マイクロフォン１０の設置位置での騒音に対する騒音低減音の低減効果を高めることができる。さらに、上記制御音伝達特性モデルＨのモデル化誤差値Ｗｕと騒音低減用しきい値ｂとが考慮されていることにより、外乱に対する騒音低減制御の安定性を確保することができる。
【００４２】
また、オーディオ装置１５の作動時には、上記マイク信号ｕからオーディオ装置１５のオーディオ信号ｏが除去され、このことで、マイクロフォン１０により検出される音のうち、低減対象である騒音からオーディオ音が除外される。この場合には、スピーカ信号ｙの設定の前にオーディオ信号ｏが除去されることにより、マイクロフォン１０がオーディオ音を検出していない状態にあると見做されるので、上記スピーカ１２の騒音低減音がオーディオ音に影響を及ぼすという事態は、各マイクロフォン１０の設置位置においてのみ回避することができる。
【００４３】
（実施例２）
図９は、本発明の実施例２に係る車両騒音低減装置の基本構成を概略的に示し、この車両騒音低減装置では、オーディオ装置１５のオーディオ信号ｏは、ＣＰＵ１６における演算ブロック４０の出力側で除去される。
【００４４】
すなわち、本発明の特徴として、上記実施例１のマイク信号ｕに代えて、ＣＰＵ１６のスピーカ信号ｙからオーディオ装置１５のオーディオ信号ｏを除去するようになっている。
【００４５】
具体的には、図１０に詳示するように、上記演算ブロック４０の出力段には、上記実施例１と同じ構成でかつ同じ数（ｌ個）のモデルフィルタ２６と、これらモデルフィルタ２６から出力されたオーディオ信号ｏに対し該演算ブロック４０と同じ定数Ｋを用いて同じ演算処理を施すオーディオ用演算ブロック４１と、このオーディオ用演算ブロック４１が出力したオーディオ信号ｏを上記スピーカ信号ｙから除去する減算器２９とが設けられている。
【００４６】
したがって、本実施例２によれば、上記実施例１と異なり、マイク信号ｕに代えて、スピーカ信号ｙからオーディオ信号ｏが除去される。この場合には、スピーカ信号ｙの設定の後にオーディオ信号ｏが除去されることにより、スピーカ１２の低減音からオーディオ音に関係する音が除外されるので、該スピーカ１２の騒音低減音がオーディオ音に影響を及ぼすという事態は、車室２内全体において回避することができる。
【００４７】
（実施例３）
図１１は、本発明の実施例３に係る車両騒音低減装置の基本構成を概略的に示し、この車両騒音低減装置では、上記実施例１のものと実施例２のものとが切換可能に接続されている。
【００４８】
つまり、本発明の特徴として、ＣＰＵ１６は、図１２に詳しく示すように、マイクロフォン１０のマイク信号ｕに加えてスピーカ信号ｙからもオーディオ装置１５のオーディオ信号ｏを除去可能に構成されている。また、上記オーディオ装置１５のオーディオ信号ｏをマイク信号ｕ又はスピーカ信号ｙから選択的に除去するために切り換える切換手段としての切換スイッチ３０が設けられている。そして、この切換スイッチ３０には、車室２内で手動入力可能なマニュアルスイッチ３１が連設されていて、このマニュアルスイッチ３１の出力によりオーディオ信号ｏがマイク信号ｕ又はスピーカ信号ｙから選択的に除去されるようになっている。尚、本実施例３のその他の構成は上記実施例１及び実施例２と同じであるので、同じ部分には同じ符号を付して示し、その説明は省略する。
【００４９】
よって、本実施例３によれば、オーディオ信号ｏを、マイク信号ｕからでもスピーカ信号ｙからでも切り換えて除去できるので、マイクロフォン１０の設置位置に限って騒音低減音のオーディオ音への影響が回避されるモードと、車室２内全体において騒音低減音のオーディオ音への影響が回避されるモードとを、乗員が任意に切り換えることができる。
【００５０】
（実施例４）
図１３は、本発明の実施例４に係る車両騒音低減装置の基本構成を概略的に示し、この低減装置では、オーディオ装置１５の作動中にはマイク信号ｕがＣＰＵ１６の演算ブロック４０に入力されるのを停止するようになっている。
【００５１】
すなわち、本発明の特徴として、上記オーディオ装置１５の作動時にはＣＰＵ１６によるスピーカ信号ｙの設定が中止されるようになされている。具体的には、各マイクロフォン１０とＣＰＵ１６の演算ブロック４０との間に、マイク信号ｕが演算ブロック４０に入力されるのをオンオフ切換作動するオンオフスイッチ３２が介設されている。また、このオンオフスイッチ３２はオーディオ装置１５のオーディオ信号ｏが入力可能に接続されており、該オーディオ信号ｏが入力されたときに上記マイク信号ｕの演算ブロック４０への入力をオフするようになっている。
【００５２】
したがって、本実施例４によると、オーディオ装置１５の作動時には、ＣＰＵ１６におけるスピーカ信号ｙの設定自体が停止されるので、オーディオ装置１５の非作動時にはロードノイズ等の騒音を効果的に低減することができる一方、オーディオ装置１５の作動時には騒音低減音に起因するオーディオ音への悪影響を完全に回避することができる。
【００５３】
【発明の効果】
以上説明したように、請求項１又は５の発明によれば、オーディオ装置を有する車室内に設置されて該設置位置における騒音源からの騒音を検出する騒音検出手段の出力信号に基づき、騒音源及び騒音検出手段間の騒音伝達特性を低下させて乗員に対する騒音を低減するための制御信号を設定し、この制御信号を低減音発生手段に出力して騒音低減音を発生させるようにした車両騒音低減装置において、低減音発生手段及び騒音検出手段間の音の伝達特性に関して予め行われた測定結果に基づいてモデル化してなる制御音伝達特性モデルＨと、上記測定結果と上記制御音伝達特性モデルとの差の絶対値に基づいて予め定められたモデル化誤差値Ｗｕと、騒音伝達特性Ｇと、該騒音伝達特性の低減騒音レベルを示すために予め定められた騒音低減用しきい値ｂと、定数Ｋとが、騒音低減制御を安定させる条件式を満たすように定数Ｋを求めて、該定数Ｋと上記騒音検出手段の出力信号とに基づいて制御信号を設定するとともに、上記オーディオ装置の作動時には、騒音検出手段の出力信号からオーディオ装置のオーディオ信号を除去するようにしたので、良好な騒音源信号を得ることが困難なロードノイズ等の騒音を安定して低減できる一方、車載オーディオ装置からのオーディオ音への悪影響を騒音検出手段の設置位置において回避することができる。
【００５４】
請求項２又は６の発明によれば、上記騒音検出手段の出力信号に代えて制御信号からオーディオ装置のオーディオ信号を除去するようにしたので、低減音発生手段の騒音低減音からオーディオ音に関係する音を一切除外でき、該低減音発生手段の騒音低減音がオーディオ音に影響を及ぼすという事態を車室内全体において回避することができる。
【００５５】
請求項３又は７の発明によれば、上記騒音検出手段の出力信号に加えて制御信号からもオーディオ装置のオーディオ信号を除去可能に構成し、上記オーディオ装置のオーディオ信号を騒音検出手段の出力信号又は制御信号から選択的に除去するために切り換えるようにしたので、上記騒音検出手段の設置位置に限って騒音低減音のオーディオ音への影響を回避できる効果と、車室内全体において騒音低減音のオーディオ音への影響を回避できる効果とを、乗員により任意に切り換えることができる。
【００５６】
請求項４又は８の発明によれば、オーディオ装置の作動時には上記制御信号の設定を中止するようにしたので、上記オーディオ装置の非作動時にはロードノイズ等の騒音を効果的に低減することができる一方、オーディオ装置の作動時には騒音低減音がオーディオ音に悪影響を及ぼすという事態を回避することができる。
【図面の簡単な説明】
【図１】本発明の実施例１に係る車両騒音低減装置においてスピーカ信号を設定するために定められるモデル化誤差値及び騒音低減用しきい値をそれぞれ示す特性図である。
【図２】車両騒音低減装置等の各機器の配置構成を概略的に示す平面図である。
【図３】マイクロフォン及びスピーカの設置位置を示す側面図である。
【図４】コントローラの構成を示すブロック図である。
【図５】コントローラにおけるＣＰＵの構成を示すブロック図である。
【図６】ＣＰＵにおける演算ブロックの構成を示す行列式の図である。
【図７】ＣＰＵにおけるスピーカ信号の設定処理動作を示すフローチャート図である。
【図８】車両騒音低減装置の基本構成を示す概略図である。
【図９】本発明の実施例２に係る車両騒音低減装置の基本構成を示す図８相当図である。
【図１０】コントローラにおけるＣＰＵの構成を示す図５相当図である。
【図１１】本発明の実施例３に係る車両騒音低減装置の基本構成を示す図８相当図である。
【図１２】コントローラにおけるＣＰＵの構成を示す図５相当図である。
【図１３】本発明の実施例３に係る車両騒音低減装置のコントローラにおけるＣＰＵの構成を示す図５相当図である。
【符号の説明】
２ 車室
１０ マイクロフォン（騒音検出手段）
１２ スピーカ（低減音発生手段）
１５ オーディオ装置
１６ ＣＰＵ（制御手段）
３０ 切換スイッチ（切換手段）
ｕ マイク信号（出力信号）
ｙ スピーカ信号（制御信号）
ｏ オーディオ信号
Ｓ 騒音源
Ｈ 制御音伝達特性モデル
Ｗｕ モデル化誤差値
Ｇ 騒音伝達特性
ｂ 騒音低減用しきい値
Ｋ 定数 [0001]
[Industrial application fields]
The present invention relates to a device for reducing noise such as road noise that occurs when a vehicle travels by noise reduction sound, and a method for setting a control signal for generating the noise reduction sound, and more particularly to audio sound of an in-vehicle audio device. It relates to measures to avoid adverse effects.
[0002]
[Prior art]
In this type of noise reduction device, for example, as known in Japanese Utility Model Publication No. 61-1739, when the engine noise as noise is reduced, the control means generates a noise source signal (reference signal) related to the engine noise. A control signal is generated on the basis of this, and the control signal is input to the shaker, and the shaker generates an inverted sound having an opposite phase and the same amplitude as the engine noise, while the engine noise should be reduced. An acceleration sensor for detecting a synthesized sound of engine noise and the inverted sound from the shaker is installed, and the control means performs gain adjustment and phase adjustment of the control signal so that the output signal of the acceleration sensor is reduced. There is something that was made. In this case, in order to avoid the possibility that an excessive excitation force is generated in the shaker and the safety is impaired when the output signal of the acceleration sensor no longer converges due to disturbance. When the value of the output signal exceeds a predetermined value, the gain adjustment and phase adjustment of the control signal are performed again.
[0003]
On the other hand, in the application filed earlier by the present applicant (Japanese Patent Laid-Open No. 5-232969), for example, in order to reduce engine noise at a predetermined location, anti-noise is generated that reduces engine noise at the predetermined location. And a control means for correcting the output signal of the microphone based on the engine noise period and the sound transmission characteristics between the microphone and the speaker. Anti-noise is generated at the speaker by the control signal of the means. In this case, since a control signal having good coherence with engine noise can be easily obtained by using the output signal of the microphone, the amount of calculation in the above control means is reduced to the conventional LMS (Least Mean Square Method [= minimum). Square method]) There is an advantage that the calculation time can be greatly shortened by reducing it to a fraction of the algorithm or less.
[0004]
These are all noise reduction devices by so-called feedforward control that reduce noise using noise source signals with good coherence with noise, and can effectively reduce the output signal of the noise detection means. It is said that.
[0005]
[Problems to be solved by the invention]
By the way, when the noise source cannot be specified, such as road noise caused by running of the vehicle, and the noise changes irregularly, the noise reduction device using the feedforward control has a good noise source signal. It is difficult to obtain noise, and it is difficult to reduce noise.
[0006]
Therefore, in order to reduce the noise as described above, there is no choice but to feedback control the control signal so that the output signal of the noise detection means becomes small. There is a problem that it becomes unstable when the characteristics change.
[0007]
On the other hand, when such noise reduction control is considered from another viewpoint, it cannot be said that all the sounds detected by the noise detection means are noises to be reduced. That is, for example, an audio sound from an audio device mounted on a vehicle is considered to be one of important elements that brings comfort and a refreshing feeling to an occupant today. Nevertheless, if the feedback control for noise reduction is pushed forward to make it more effective, even the audio sound is reduced, and there is a problem that the effect of the audio sound cannot be obtained.
[0008]
The present invention has been made in view of the above, and an object thereof is to stably reduce noise such as road noise, which is difficult to obtain a good noise source signal, while reducing audio noise from a vehicle-mounted audio device. It is to be able to avoid adverse effects.
[0009]
[Means for Solving the Problems]
In order to achieve the above object, the invention of claim 1A control sound transfer characteristic model H that is modeled based on a measurement result performed in advance with respect to a sound transfer characteristic between the reduced sound generation means and the noise detection means, and a difference between the measurement result and the control sound transfer characteristic model. A modeling error value Wu determined in advance based on the absolute value, a noise transfer characteristic G, a noise reduction threshold value b predetermined to indicate a reduced noise level of the noise transfer characteristic, a constant K, Calculates a constant K so as to satisfy the conditional expression for stabilizing the noise reduction control, and sets the control signal based on the constant K and the output signal of the noise detection means.On the other hand, by removing the audio signal of the audio device from the output signal of the noise detection means, it is possible to avoid adverse effects on the audio sound while effectively reducing noise such as road noise with stable control.
[0010]
Specifically, in the present invention, a noise detection unit that is installed in a vehicle interior having an audio device and detects noise from a noise source at the installation position, and receives the output signal of the noise detection unit, the noise source and Control means for setting a control signal for reducing noise transmission to the occupant by reducing noise transmission characteristics between the noise detection means, and a reduction sound generating means for generating noise reduction sound in response to the control signal from the control means; The vehicle noise reduction device provided with
[0011]
The audio signal of the audio apparatus is input to the control means. In addition, the control meansA control sound transfer characteristic model H that is modeled on the basis of a measurement result performed in advance with respect to a sound transfer characteristic between the reduced sound generation means and the noise detection means, and a difference between the measurement result and the control sound transfer characteristic model A modeling error value Wu that is predetermined based on the absolute value of the noise, the noise transmission characteristic G, a noise reduction threshold value b that is predetermined to indicate a reduced noise level of the noise transmission characteristic, and a constant. A constant K is obtained so that K satisfies the following expression (1), and the control signal is set based on the constant K and the output signal of the noise detection means,It is assumed that the audio signal of the audio device is removed from the output signal of the noise detection means when the audio device is in operation.
[0012]
[Equation 5]

(In formula (1), Wp = 1 / b, ‖Q‖ _∞ Is the maximum gain of the transfer function Q from the target value to the controlled variable in the closed loop, and sup _W ‖Q (jw) ‖ is defined, and ‖ · ‖ represents the norm of the matrix. )
[0013]
In the invention of claim 2, in the invention of claim 1, the control means is configured to remove the audio signal of the audio device from the control signal instead of the output signal of the noise detection means.
[0014]
In the invention of claim 3, in the invention of claim 1, the control means is configured to be able to remove the audio signal of the audio apparatus from the control signal in addition to the output signal of the noise detection means. Then, switching means for switching is provided to selectively remove the audio signal of the audio device from the output signal of the noise detection means or the control signal of the control means.
[0015]
In the invention of claim 4, based on the same premise as that of the invention of claim 1 and assuming that the audio signal of the audio device is input to the control means, the control means includes:A control sound transfer characteristic model H that is modeled on the basis of a measurement result performed in advance with respect to a sound transfer characteristic between the reduced sound generation means and the noise detection means, and a difference between the measurement result and the control sound transfer characteristic model A modeling error value Wu that is predetermined based on the absolute value of the noise, the noise transmission characteristic G, a noise reduction threshold value b that is predetermined to indicate a reduced noise level of the noise transmission characteristic, and a constant. A constant K is obtained so that K satisfies the following expression (1), and the control signal is set based on the constant K and the output signal of the noise detection means,In addition, when the audio device is in operation, the control signal setting is stopped.
[0016]
[Formula 6]

(In formula (1), Wp = 1 / b, ‖Q‖ _∞ Is the maximum gain of the transfer function Q from the target value to the controlled variable in the closed loop, and sup _W Defined by ‖Q (jw) ‖ And ‖ / ‖ represents the norm of the matrix. )
[0017]
In the fifth aspect of the present invention, noise detection means for detecting noise from the noise source at the installation position is installed in the vehicle interior having the audio device, and based on the output signal of the noise detection means, between the noise source and the noise detection means. The control signal in the vehicle noise reduction apparatus in which the control signal for reducing the noise transmission characteristic is set to reduce the noise to the occupant and the control signal is output to the reduction sound generating means to generate the noise reduction sound. As a setting method ofA control sound transfer characteristic model that is modeled on the basis of a measurement result obtained in advance with respect to the sound transfer characteristic between the reduced sound generation means and the noise detection means is determined, and the difference between the measurement result and the control sound transfer characteristic model A modeling error value is determined based on the absolute value of the noise, a noise reduction threshold for indicating the noise level of the noise transfer characteristic is determined, and the control sound transfer characteristic model H and the modeling error value Wu are determined. And the noise transmission characteristic G, the noise reduction threshold value b, and the constant K are determined so that the following equation (1) is satisfied, and based on the constant K and the output signal of the noise detection means. To set the control signal andWhen the audio apparatus is in operation, the audio signal of the audio apparatus is removed from the output signal of the noise detection means.
[0018]
[Expression 7]

(In formula (1), Wp = 1 / b, ‖Q‖ _∞ Is the maximum gain of the transfer function Q from the target value to the controlled variable in the closed loop, and sup _W ‖Q (jw) ‖ is defined, and ‖ · ‖ represents the norm of the matrix. )
[0019]
According to the invention of claim 6, in the invention of claim 5, the audio signal of the audio device is removed from the control signal instead of the output signal of the noise detecting means.
[0020]
According to the invention of claim 7, in the invention of claim 5, in addition to the output signal of the noise detection means, the audio signal of the audio device is also removed from the control signal, and the audio signal of the audio device is converted into the noise detection means. Switching to selectively remove from the output signal or control signal.
[0021]
In the invention of claim 8, based on the same premise as the invention of claim 5 above,A control sound transfer characteristic model that is modeled on the basis of a measurement result obtained in advance with respect to the sound transfer characteristic between the reduced sound generation means and the noise detection means is determined, and the difference between the measurement result and the control sound transfer characteristic model A modeling error value is determined based on the absolute value of the noise, a noise reduction threshold for indicating the noise level of the noise transfer characteristic is determined, and the control sound transfer characteristic model H and the modeling error value Wu are determined. And the noise transmission characteristic G, the noise reduction threshold value b, and the constant K are determined so that the following equation (1) is satisfied, and based on the constant K and the output signal of the noise detection means. To set the control signal andThe setting of the control signal is stopped when the audio device is operated.
[0022]
[Equation 8]

(In the formula (1), Wp = 1 / b, ‖Q‖ _∞ Is the maximum gain of the transfer function Q from the target value to the controlled variable in the closed loop, and sup _W ‖Q (jw) ‖ is defined, and ‖ · ‖ represents the norm of the matrix. )
[0023]
[Action]
In the first or fifth aspect of the present invention, noise from the noise source is detected by the noise detection means in the vehicle interior. Then, a control signal is set by the control means that receives the output signal of the noise detection means, and a noise reduction sound is generated by the reduction sound generation means that receives the control signal. Due to this noise reduction sound, the noise transmission characteristics between the noise source and the noise detection means are lowered, and thus noise to the occupant is reduced. At this time, in the control means,The constant K is determined so that the control sound transfer characteristic model H, the modeling error value Wu, the noise transfer characteristic G, the noise reduction threshold value b, and the constant K satisfy Expression (1). A control signal is set based on the output signal of the detection means.That is, by using the control sound transfer characteristic model, the noise reduction effect of the noise reduction sound at the installation position of the noise detection means is enhanced. At this time, by considering the modeling error value and the noise reduction threshold value, stability against disturbance when the control signal is feedback-controlled so as to reduce the output signal of the noise generating means is ensured.
[0024]
Also, when the audio device is in operation, the audio signal of the audio device is removed from the output signal of the noise detection means by the control means, so that, among the sounds detected by the noise detection means, the noise from the noise to be reduced is reduced to audio. Sound is excluded. In this case, since the output signal of the audio device main body is removed before the control signal is set, it is considered that the noise detecting means is not detecting the audio sound. The situation where the noise reduction sound of the means affects the audio sound is avoided at the installation position of the noise detection means.
[0025]
In the invention of claim 2 or 6, the audio signal of the audio device is removed from the control signal by the control means instead of the output signal of the noise detection means. In this case, since the audio signal is removed after the control signal is set, the sound related to the audio sound is excluded from the noise reduction sound of the reduction sound generating means, so that the noise reduction sound affects the audio sound. This situation is avoided in the entire vehicle interior.
[0026]
In the invention of claim 3 or 7, the audio signal of the audio device is switched and removed from the output signal of the noise detection means or from the control signal, so that the audio of the noise reduction sound is limited to the installation position of the noise detection means. The occupant can arbitrarily switch between the action of avoiding the influence on the sound and the action of avoiding the influence of the noise reduction sound on the audio sound in the entire vehicle interior.
[0027]
In the invention of claim 4 or 8, the setting of the control signal is stopped when the audio apparatus is operated. Accordingly, road noise and other noises are effectively reduced when the audio device is not operating, while the situation where the noise reducing sound adversely affects the audio sound when the audio device is operating is completely avoided.
[0028]
【Example】
Embodiments of the present invention will be described below with reference to the drawings.
[0029]
Example 1
2 and 3 schematically show the overall configuration of the noise reduction device according to the first embodiment of the present invention. This noise reduction device is a load for passengers in each seat in an automobile (vehicle) equipped with an audio device. This is for reducing noise of several tens to several hundreds Hz, such as noise, without using a noise source signal (reference signal).
[0030]
In the figures, 1 is a vehicle body, 2 is a passenger compartment located in the center of the front and rear of the vehicle body 1, 3 and 4 are front and rear doors that open and close the passenger compartment 2, The front seat 5 (the left seat in FIGS. 2 and 3) is the right driver seat 5a and the left passenger seat 5b, and the rear seat 6 (see FIGS. 2 and 3). Each right seat) is composed of two left and right rear seats 6a, 6a. 7 is a steering wheel installed in front of the driver's seat 5a.
[0031]
In the first embodiment, the left or right side of each headrest 8 in the driver seat 5a, the passenger seat 5b, and the rear seat 6a (the position close to either the left or right ear of the seated occupant) is set as the noise reduction location. In each of these locations, four microphones 10 as noise detection means are attached in total, and the sound in the passenger compartment 2 is detected near the occupant's ear by these microphones 10, 10,. . On the other hand, reduced sound generation for generating noise reducing sound in the vehicle compartment 2 is formed on the inner surface of the vehicle compartment 2 of the front doors 3 and 3 on the left and right sides of the vehicle body 1 and on the left and right sides of the package tray 11 at the rear end of the vehicle compartment 2 .. Are arranged as means, and these four speakers 12, 12,... Are also used as the audio device 15. An audio device 15 is installed at the center of the instrument panel 13 on the left and right.
[0032]
The speakers 12, 12,... And the microphones 10, 10,... Are connected to a controller 14 as control means disposed in the instrument panel 13 on the passenger seat 5b side, for example. A synthesized sound of the noise in the passenger compartment 2 and the noise-reduced sound emitted from each speaker 12 is detected by each microphone 10, and a microphone signal u output from the microphone 10 is input to the controller 14, By outputting a speaker signal y as a control signal having a phase opposite to that of the microphone signal u to the speaker 12, noise reduction sound from each speaker 12 is caused to interfere with noise at the position of each microphone 10, and each microphone 10 The detected sound is reduced.
[0033]
As shown in detail in FIG. 4, the controller 14 has a CPU 16 (central processing unit) as a control unit that outputs a speaker signal y for reducing the microphone signal u by digital signal processing. In the input stage, a microphone amplifier 17 that amplifies the microphone signal u from each microphone 10, and a low-pass filter 18 that waves a low frequency portion (for example, 500 to 1000 Hz or less) of the amplified microphone signal u are filtered. The analog microphone signal u is converted to a digital signal, and an A / D converter 19 is sequentially connected in the same number as the number of microphones 10, 10,. On the other hand, at the output stage of the CPU 16, a D / A converter 20 that converts the digital speaker signal y into an analog signal, a low-pass filter 21 that ripples a low-frequency portion of the speaker signal y converted into an analog signal, A speaker amplifier 22 that amplifies the loudspeaker speaker signal y is sequentially connected in the same number as the number of speakers 12, 12,. The input stage of the CPU 16 includes a microphone amplifier 23 that amplifies the audio signal o from the audio device main body 15 and a low-pass filter that ripples a low frequency portion (eg, 500 to 1000 Hz or less) of the amplified audio signal o. 24 and an A / D conversion unit 25 that converts the analog audio signal o that has been waved into a digital signal are connected in order. The operations of the arithmetic processing block 16, the A / D converters 19 and 25, and the D / A converter 20 are performed in synchronization with each other by a sampling period signal generated by a sampling clock generator (not shown).
[0034]
The configuration of the CPU 16 will be described in more detail with reference to FIG. The microphone signal u is a column vector composed of m (here, m = 4) microphone signals u1 to um. Basically, based on these microphone signals u1 to um, the calculation block 40 provided in the CPU 16 A speaker signal y as a column vector composed of l (here, l = 4) speaker signals y1 to yl is calculated by a constant K. The constant K is a matrix composed of four matrix components A to D as shown in FIG. 6, and the calculation process is performed as shown in the flowchart of FIG. That is, after inputting the microphone signal u in step S1, in step S2, the value obtained by multiplying the previous microphone signal x by the C component and the value obtained by multiplying the current microphone signal u by the D component are added. Then, the speaker signal y is obtained. In step S3, the speaker signal y is output to the speaker 12, and then the process proceeds to step S4. In this step S4, the value obtained by multiplying the previous microphone signal x by the A component and the value obtained by multiplying the current microphone signal u by the B component are added, and the previous microphone signal x is processed in the above step S2. A value to be the signal x is obtained, and then the process returns to step S1.
[0035]
As a feature of the present invention, as shown in FIG. 1A, the CPU 16 models the control sound transfer characteristic model H obtained by modeling the transfer characteristic of the noise reduction sound that reaches the microphone 10 from the speaker 12. An error value Wu (Uncertainty Weight) and a noise reduction purpose determined to indicate the noise level to be reduced of the noise transfer characteristic G entering the microphone 10 from the noise source S as shown in FIG. The speaker signal y is set based on a constant K determined by a threshold value b (Performance Objective) (for simplification of explanation, the speakers 12, 12,... And the microphones 10, 10). ,... Are each one, and each of the audio device 15 and the noise reduction device has a dedicated speaker). Further, when the audio device 15 is operated, the audio signal o of the audio device 15 is removed from the microphone signal u.
[0036]
Specifically, in the control sound transfer characteristic model H, after the speaker signal y is output from the CPU 16, the speakers 12, 12,. The sound transmission characteristics until the change in sound is detected by the microphone 10 and the microphone signal u is input to the CPU 16 are modeled based on the actual measurement result H ′. In consideration of the modeling error at that time, the modeling error value Wu for the absolute value | H−H ′ | of the error amount is determined when the constant K of the calculation block 40 is determined. On the other hand, a threshold b for noise reduction is determined for the noise transfer characteristic G, and the peak of the closed loop transfer function including the controller 14 is intensively dropped below the line b. That is, if the noise reduction weight Wp (Performance Weight) is Wp = 1 / b, for all frequencies,
[Equation 9]

That means
[Expression 10]

The relational expression of The determination of the modeling error value Wu and the noise reduction threshold value b determines the constant K and sets the speaker signal y.
[0037]
At this time, the modeling error value Wu and the noise reduction threshold value b cannot be determined independently of each other. For example, if the modeling error value Wu is decreased to increase the noise reduction effect, the noise reduction control becomes unstable because the acoustic characteristics in the passenger compartment 2 are easily changed. If it is increased, the noise reduction threshold value b cannot be lowered this time. Therefore, the following equation is known as a mixed sensitivity problem in the case of feedback characteristics in the H∞ control theory, which is one of robust controls.(1)Modeling error value such thatWuAnd noise reduction thresholdbTo select the constant K.
[0038]
[Expression 11]

(In the formula (1), Wp = 1 / b, ‖Q‖ _∞ Is the maximum gain of the transfer function Q from the target value to the controlled variable in the closed loop, and sup _W ‖Q (jw) ‖ is defined, and ‖ · ‖ represents the norm of the matrix. )
[0039]
Next, a circuit for setting the speaker signal y by removing the audio signal o of the audio device 15 from the microphone signal u will be described in more detail with reference to FIGS.
[0040]
As shown in principle in FIG. 8, the input stage of the calculation block 40 in the CPU 16 is based on a control sound transfer characteristic model H1 that models the sound transfer characteristic between the speaker 12 and the microphone 10, respectively. A model filter 26 for adjusting the phase and gain of the audio signal o output to the speaker 12 and a subtractor 27 for removing the audio signal o output from the model filter 26 from the microphone signal u are provided. Specifically, as shown in FIG. 5, the model filters 26 are arranged by the number (m) of the microphone signals u, and each model filter 26 has the same number (l) of model filters as the speaker signals y. 26a, 26a,... Are connected in parallel. Each of the model filters 26a includes control sound transfer characteristic models H1 to H1 in which sound transfer characteristics until the noise reduction sound of the speaker 12 reaches the microphone 10 are modeled. That is, for the first microphone 10, model filters 26a, 26a,... Composed of control sound transfer characteristic models H11, H21,..., H11 between the microphone 10 and the l speakers 12, 12,. Therefore, for the m-th microphone 10, the model filter 26 provided is a model composed of control sound transfer characteristic models H1m, H2m,..., Hlm between the microphone 10 and the l speakers 12, 12,. A model filter 26 having filters 26a, 26a,... Further, each model filter 26 is provided with an adder 26b that adds the output values of the model filters 26a, 26a,... And outputs the sum to the subtractor 27.
[0041]
Therefore, according to the first embodiment, noise from the noise source S is detected by the microphone 10 in the driver seat 5a, the passenger seat 5b, and the rear seat 6a in the passenger compartment 2. The CPU 16 that has received the microphone signal u sets the speaker signal y, and the speaker 12 that has received the speaker signal y generates noise reduction sound. Due to this noise reduction sound, the noise transfer characteristic G between the noise source S and the microphone 10 is lowered, and thereby the noise d to the occupant is reduced. At this time, by using the control sound transfer characteristic model H in the CPU 16, it is possible to enhance the noise reduction sound reduction effect with respect to the noise at the installation position of the microphone 10. Furthermore, since the modeling error value Wu of the control sound transfer characteristic model H and the noise reduction threshold value b are taken into account, the stability of the noise reduction control against disturbance can be ensured.
[0042]
Further, when the audio device 15 is operated, the audio signal o of the audio device 15 is removed from the microphone signal u, and thereby, the audio sound is excluded from the noise to be reduced among the sounds detected by the microphone 10. The In this case, since the audio signal o is removed before the speaker signal y is set, it is considered that the microphone 10 is not detecting the audio sound. Can affect the audio sound only at the installation position of each microphone 10.
[0043]
(Example 2)
FIG. 9 schematically shows a basic configuration of the vehicle noise reduction device according to the second embodiment of the present invention. In this vehicle noise reduction device, the audio signal o of the audio device 15 is output on the output side of the calculation block 40 in the CPU 16. Removed.
[0044]
That is, as a feature of the present invention, the audio signal o of the audio device 15 is removed from the speaker signal y of the CPU 16 instead of the microphone signal u of the first embodiment.
[0045]
Specifically, as shown in detail in FIG. 10, the output stage of the arithmetic block 40 includes the same number (l) of model filters 26 having the same configuration as in the first embodiment and the model filters 26. An audio calculation block 41 that performs the same calculation process on the output audio signal o using the same constant K as the calculation block 40, and the audio signal o output by the audio calculation block 41 is removed from the speaker signal y. And a subtractor 29 is provided.
[0046]
Therefore, according to the second embodiment, unlike the first embodiment, the audio signal o is removed from the speaker signal y instead of the microphone signal u. In this case, since the audio signal o is removed after the setting of the speaker signal y, the sound related to the audio sound is excluded from the reduced sound of the speaker 12, so that the noise reduced sound of the speaker 12 is the audio sound. The situation that affects the vehicle interior 2 can be avoided in the entire cabin 2.
[0047]
(Example 3)
FIG. 11 schematically shows a basic configuration of a vehicle noise reduction device according to a third embodiment of the present invention. In this vehicle noise reduction device, the one of the first embodiment and the second embodiment are switchably connected. Has been.
[0048]
That is, as a feature of the present invention, the CPU 16 is configured to be able to remove the audio signal o of the audio device 15 from the speaker signal y in addition to the microphone signal u of the microphone 10, as shown in detail in FIG. Further, a changeover switch 30 is provided as a changeover means for changing over to selectively remove the audio signal o of the audio device 15 from the microphone signal u or the speaker signal y. A manual switch 31 that can be manually input in the passenger compartment 2 is connected to the changeover switch 30, and the audio signal o is selectively selected from the microphone signal u or the speaker signal y by the output of the manual switch 31. It is supposed to be removed. In addition, since the other structure of this Example 3 is the same as the said Example 1 and Example 2, it attaches | subjects and shows the same code | symbol to the same part, The description is abbreviate | omitted.
[0049]
Therefore, according to the third embodiment, since the audio signal o can be switched and removed from the microphone signal u or the speaker signal y, the influence of the noise reduction sound on the audio sound is avoided only in the installation position of the microphone 10. The passenger can arbitrarily switch between the mode to be performed and the mode in which the influence of the noise reduction sound on the audio sound is avoided in the entire cabin 2.
[0050]
(Example 4)
FIG. 13 schematically shows the basic configuration of a vehicle noise reduction device according to Embodiment 4 of the present invention. In this reduction device, a microphone signal u is input to the calculation block 40 of the CPU 16 while the audio device 15 is in operation. Is supposed to stop.
[0051]
That is, as a feature of the present invention, the setting of the speaker signal y by the CPU 16 is stopped when the audio device 15 is operated. More specifically, an on / off switch 32 is provided between each microphone 10 and the calculation block 40 of the CPU 16 for switching on / off the input of the microphone signal u to the calculation block 40. The on / off switch 32 is connected so that the audio signal o of the audio device 15 can be input. When the audio signal o is input, the input of the microphone signal u to the calculation block 40 is turned off. ing.
[0052]
Therefore, according to the fourth embodiment, since the setting of the speaker signal y in the CPU 16 is stopped when the audio device 15 is operated, noise such as road noise can be effectively reduced when the audio device 15 is not operated. On the other hand, when the audio device 15 is operated, it is possible to completely avoid an adverse effect on the audio sound caused by the noise reduction sound.
[0053]
【The invention's effect】
As described above, according to the first or fifth aspect of the present invention, the noise source is installed on the basis of the output signal of the noise detecting means that is installed in the vehicle interior having the audio device and detects noise from the noise source at the installation position. Vehicle noise in which a noise transmission characteristic between the noise detection means and the control signal for reducing the noise to the occupant is set and the control signal is output to the reduction sound generation means to generate the noise reduction sound. In the reduction device,A control sound transfer characteristic model H that is modeled based on a measurement result performed in advance with respect to a sound transfer characteristic between the reduced sound generation means and the noise detection means, and a difference between the measurement result and the control sound transfer characteristic model. Predetermined modeling error value Wu based on absolute value, noise transfer characteristic G, noise reduction threshold value b predetermined to indicate a reduced noise level of the noise transfer characteristic, constant K, Calculates a constant K so as to satisfy the conditional expression for stabilizing the noise reduction control, sets the control signal based on the constant K and the output signal of the noise detection means,During the operation of the audio device, the audio signal of the audio device is removed from the output signal of the noise detection means, so that it is possible to stably reduce noise such as road noise which is difficult to obtain a good noise source signal. Thus, an adverse effect on the audio sound from the in-vehicle audio device can be avoided at the installation position of the noise detection means.
[0054]
According to the invention of claim 2 or 6, since the audio signal of the audio device is removed from the control signal instead of the output signal of the noise detection means, the noise reduction sound of the reduction sound generation means is related to the audio sound. Therefore, it is possible to avoid the situation in which the noise reduction sound generated by the reduction sound generating means affects the audio sound.
[0055]
According to the third or seventh aspect of the present invention, the audio signal of the audio device can be removed from the control signal in addition to the output signal of the noise detection means, and the audio signal of the audio device is output as the output signal of the noise detection means. Alternatively, since the switching is performed to selectively remove the control signal from the control signal, it is possible to avoid the influence of the noise reduction sound on the audio sound only in the installation position of the noise detection means, and the noise reduction sound in the entire vehicle interior. The effect of avoiding the influence on the audio sound can be arbitrarily switched by the occupant.
[0056]
According to the invention of claim 4 or 8TheSince the setting of the control signal is stopped when the audio device is in operation, noise such as road noise can be effectively reduced when the audio device is inactive, while noise reduction sound is generated when the audio device is in operation. It is possible to avoid a situation where the audio sound is adversely affected.
[Brief description of the drawings]
FIG. 1 is a characteristic diagram showing a modeling error value and a noise reduction threshold value that are determined for setting a speaker signal in the vehicle noise reduction device according to the first embodiment of the present invention.
FIG. 2 is a plan view schematically showing an arrangement configuration of devices such as a vehicle noise reduction device.
FIG. 3 is a side view showing the installation positions of a microphone and a speaker.
FIG. 4 is a block diagram illustrating a configuration of a controller.
FIG. 5 is a block diagram illustrating a configuration of a CPU in a controller.
FIG. 6 is a determinant diagram showing a configuration of a calculation block in a CPU.
FIG. 7 is a flowchart showing a speaker signal setting processing operation in a CPU.
FIG. 8 is a schematic diagram showing a basic configuration of a vehicle noise reduction device.
FIG. 9 is a view corresponding to FIG. 8 showing a basic configuration of a vehicle noise reduction apparatus according to Embodiment 2 of the present invention.
FIG. 10 is a view corresponding to FIG. 5 showing a configuration of a CPU in the controller.
FIG. 11 is a view corresponding to FIG. 8 showing a basic configuration of a vehicle noise reduction apparatus according to Embodiment 3 of the present invention.
12 is a view corresponding to FIG. 5 showing a configuration of a CPU in the controller.
FIG. 13 is a view corresponding to FIG. 5 and showing the configuration of the CPU in the controller of the vehicle noise reduction device according to Embodiment 3 of the present invention.
[Explanation of symbols]
2 compartment
10 Microphone (noise detection means)
12 Speaker (reducing sound generation means)
15 Audio equipment
16 CPU (control means)
30 selector switch (switching means)
u Microphone signal (output signal)
y Speaker signal (control signal)
o Audio signal
S Noise source
H Control sound transfer characteristic model
Wu modeling error value
G Noise transmission characteristics
b Noise reduction threshold
K    constant

Claims (8)

A noise detection means installed in a vehicle interior having an audio device and detecting noise from a noise source at the installation position;
Control means for receiving an output signal of the noise detection means and setting a control signal for reducing noise to an occupant by reducing noise transmission characteristics between the noise source and the noise detection means;
In a vehicle noise reduction device comprising a reduction sound generation means for generating a noise reduction sound in response to a control signal from the control means,
The audio signal of the audio device is input to the control means,
The control means includes a control sound transfer characteristic model H that is modeled based on a measurement result performed in advance with respect to a sound transfer characteristic between the reduced sound generation means and the noise detection means, the measurement result, and the control sound transmission. A noise reduction threshold predetermined to indicate the modeling error value Wu determined in advance based on the absolute value of the difference from the characteristic model, the noise transmission characteristic G, and the reduced noise level of the noise transmission characteristic. The constant K is obtained so that the value b and the constant K satisfy the following expression (1), the control signal is set based on the constant K and the output signal of the noise detecting means, and the audio device A vehicle noise reduction device configured to remove the audio signal of the audio device from the output signal of the noise detection means during operation.

(In the formula (1), Wp = 1 / b, ‖Q‖ ∞ is defined a maximum gain of the transfer function Q from the target value in the closed loop to control the amount by sup _W ‖Q (jw) || (‖ And ‖ represent the norm of the matrix.) The vehicle noise reduction device according to claim 1,
The vehicle noise reduction device, wherein the control means is configured to remove the audio signal of the audio device from the control signal instead of the output signal of the noise detection means. The vehicle noise reduction device according to claim 1,
The control means is configured to be able to remove the audio signal of the audio device from the control signal in addition to the output signal of the noise detection means,
A vehicle noise reduction apparatus comprising switching means for selectively removing an audio signal of the audio apparatus from an output signal of the noise detection means or a control signal of the control means. A noise detection means installed in a vehicle interior having an audio device and detecting noise from a noise source at the installation position;
Control means for receiving an output signal of the noise detection means and setting a control signal for reducing noise to an occupant by reducing noise transmission characteristics between the noise source and the noise detection means;
In a vehicle noise reduction device comprising a reduction sound generating means for generating a noise reduction sound in response to a control signal from the control means,
The audio signal of the audio device is input to the control means,
The control means includes a control sound transfer characteristic model H that is modeled based on a measurement result performed in advance with respect to a sound transfer characteristic between the reduced sound generation means and the noise detection means, the measurement result, and the control sound transmission. Based on the absolute value of the difference from the characteristic model, a predetermined modeling error value W u, the noise transmission characteristic G, and a noise reduction characteristic predetermined to indicate the reduced noise level of the noise transmission characteristic are used. The constant K is obtained so that the threshold value b and the constant K satisfy the following formula (1), the control signal is set based on the constant K and the output signal of the noise detection means, and the audio A vehicle noise reduction device configured to stop the setting of the control signal when the device is in operation.

(In the formula (1), Wp = 1 / b, ‖Q‖ ∞ is defined a maximum gain of the transfer function Q from the target value in the closed loop to control the amount by sup _W ‖Q (jw) || (‖ And ‖ represent the norm of the matrix.) A noise detection means for detecting noise from a noise source at the installation position is installed in the vehicle interior having the audio device,
Based on the output signal of the noise detection means, a control signal for reducing noise to the passenger by reducing the noise transmission characteristics between the noise source and the noise detection means,
A method of setting the control signal in a vehicle noise reduction device that generates the noise reduction sound by outputting the control signal to a reduction sound generation means,
A control sound transfer characteristic model that is modeled on the basis of a measurement result obtained in advance with respect to the sound transfer characteristic between the reduced sound generation means and the noise detection means is determined, and the difference between the measurement result and the control sound transfer characteristic model A modeling error value is determined based on the absolute value of the noise, a noise reduction threshold value for indicating the noise level of the noise transfer characteristic is determined, and the control sound transfer characteristic model H and the modeling error value Wu are determined. The noise transfer characteristic G, the noise reduction threshold value b, and the constant K are determined so that the following equation (1) is satisfied, and based on the constant K and the output signal of the noise detection means. And setting the control signal and removing the audio signal of the audio device from the output signal of the noise detecting means when the audio device is in operation.

(In the formula (1), Wp = 1 / b, ‖Q‖ ∞ is defined a maximum gain of the transfer function Q from the target value in the closed loop to control the amount by sup _W ‖Q (jw) || (‖ And ‖ represent the norm of the matrix.) The control signal setting method according to claim 5, wherein
A control signal setting method, wherein the audio signal of the audio device is removed from the control signal instead of the output signal of the noise detection means. The control signal setting method according to claim 5, wherein
In addition to the output signal of the noise detection means, the audio signal of the audio device is also removed from the control signal,
A control signal setting method, wherein the audio signal of the audio device is switched in order to selectively remove the audio signal from the output signal or control signal of the noise detecting means. A noise detection means for detecting noise from a noise source at the installation position is installed in the vehicle interior having the audio device,
Based on the output signal of the noise detection means, a control signal for reducing noise to the passenger by reducing the noise transmission characteristics between the noise source and the noise detection means,
A method of setting the control signal in a vehicle noise reduction device that generates the noise reduction sound by outputting the control signal to a reduction sound generation means,
A control sound transfer characteristic model that is modeled on the basis of a measurement result obtained in advance with respect to the sound transfer characteristic between the reduced sound generation means and the noise detection means is determined, and the difference between the measurement result and the control sound transfer characteristic model A modeling error value is determined based on the absolute value of the noise, a noise reduction threshold value for indicating the noise level of the noise transfer characteristic is determined, and the control sound transfer characteristic model H and the modeling error value Wu are determined. The noise transfer characteristic G, the noise reduction threshold value b, and the constant K are determined so that the following equation (1) is satisfied, and based on the constant K and the output signal of the noise detection means. And setting the control signal, and stopping the setting of the control signal when the audio apparatus is in operation.

(In the formula (1), Wp = 1 / b, ‖Q‖ ∞ is defined a maximum gain of the transfer function Q from the target value in the closed loop to control the amount by sup _W ‖Q (jw) || (‖ And ‖ represent the norm of the matrix.)

Images