JP3910643B2 - Integrated laser / infrared forward monitoring optical device - Google Patents

Description

図２２は、三相モータ、ＰＷＭ増幅器サーボ・システム回路２２００の回路図である。三相モータ、ＰＷＭ増幅器サーボ・システム回路２２００は、上側励振トランジスタＴ１及び下側励振トランジスタＴ４を含む位相励振器Ａ、上側励振トランジスタＴ２及び下側励振トランジスタＴ５を含む位相励振器Ｂ、及び上側２５励振トランジスタＴ３及び下側励振トランジスタＴ６を含む位相励振器Ｃを有するＰＷＭ増幅器を採用する。重要なのは、励振トランジスタがＩＧＢＴであることに注意することである。しかしながら、やはり、当業者が認めるように、ＭＯＳＦＥＴ及びバイポーラ・トランジスタのような他の固体スイッチを採用することもできる。三相モータ、ＰＷＭ増幅器サーボ・システム回路２２００もまた制御論理２２０５を採用し、この制御論理は次の励振トランジスタ対、すなわち、Ｔ１とＴ５、Ｔ１とＴ６、Ｔ２とＴ６、Ｔ２とＴ４、Ｔ３とＴ４、及びＴ３とＴ５の活性化を制御する。当業者が容易に理解するように、上に識別したトランジスタ対の活性化は、三相モータＭの３つの巻線Ｍ_A、Ｍ_B、Ｍ_Cの各々を通る電流を制御する。上に識別した励振トランジスタ対の各々を適当に活性化することによって、制御論理２２０５は、三相モータＭの軸上に定トルクを維持することができる。
制御論理２２０５は、本発明の好適実施例ではソフトウェアによって制御される。したがって、制御論理２２０５は、当業者によってホール（Ｈａｌｌ）符号又は位相符号化器帰還符号と呼ばれる一連の整流符号を受信する。各ホール符号は、制御論理２２０５に適当な励振トランジスタ制御線路１、……６を活性化させる。例えば、トランジスタＴ１及びＴ５を含むトランジスタ対を活性化する任を負うホール符号は、モータ軸の実位置にかかわらず、制御論理２２０５にトランジスタ制御線路１及び５を活性化せしめることになる。
本発明によれば、上に説明した二相モータ、ＰＷＭ増幅器サーボ故障分離プロセスのように、三相モータ、ＰＷＭ増幅器サーボ故障分離プロセスは、故障分離目的のために現存するサーボ・システム回路２２００ハードウェアを利用する。したがって、三相モータ、ＰＷＭ増幅器サーボ・システム回路２２００はまた、３つの上側電流センシング抵抗器Ｒ_A、Ｒ_B、Ｒ_C、及び下側電流センシング抵抗器Ｒ_outを含む。一般に、故障分離は、サーボ・システム回路に電源電圧Ｖ_CCの既知の量を印加し、かつ次いで上側電流センシング抵抗器Ｒ_A、Ｒ_B、Ｒ_C、及び／又は下側電流センシング抵抗器Ｒ_outを通して流れる電流の量を分析することで以て単一点故障条件を分離することによって、完遂される。三相モータ、ＰＷＭ増幅器サーボ・システム故障分離プロセスは、次の単一点故障条件、すなわち、モータ巻線開路、モータ巻線短絡線絡、モータ巻線地絡、増幅器励振トランジスタ開路、増幅器励振トランジスタ短絡、増幅器出力ワイヤボンド開路を分離することができる。
図２３Ａ及び２３Ｂは、本発明の好適実施例に従う、三相モータ、ＰＷＭ増幅器故障分離プロセスの詳細ステップを示す流れ図である。図２３Ａによれば、制御論理２２０５は、モータ巻線を通る電流の整流を制御するために本来使用されるホール符号を取り消すことによって、かつ、ブロック２３０５によって示されたように、所望励振トランジスタ対（例えば、上側励振トランジスタＴ１及び下側励振トランジスタＴ５）に相当するホール符号を発生することによって、開始する。ホール符号は、立ち代わって、ブロック２３１０に示されたように、制御論理２２０５に所望トランジスタ対に相当する適当なトランジスタ制御線路（例えば、トランジスタ制御線路１及び５）を活性化させることになる。
次に、制御論理２２０５は、ブロック２３１５によって示されたように、相当する上側電流センシング抵抗器を通して流れる電流（例えば、上側電流センシング抵抗器Ｒ_A又はＲ_B又はＲ_Cをそれぞれ通る電流Ｉ_A又はＩ_B又はＩ_C）が所定しきい値Ｔを超えるかどうか判定しなければならなず、ここでしきい値Ｔはそうでなければ電流の過大量であるものを表す。もし上側電流センシング抵抗器が電流の過大量を引き出しつつあるならば、判定ブロック２３１５から出る“ＹＥＳ”経路に従って、制御論理２２０５は、ブロック２３１７によって示されたように、ＳＨＵＴＤＯＷＮ命令を発生し、かつブロック２３２０によって示されたように適当な状態フラグをセットし、このようにして現在励振トランジスタ対（例えば、Ｔ１及びＴ５）の活性化が過電流条件を生じたことを表示することになる。
しかしながら、もし現在の励振トランジスタ対の活性化がＳＨＵＴＤＯＷＮを生じなかったならば、判定ブロック２３１５から出る“ＮＯ”経路に従って、制御論理２２０５は、判定ブロック２３２５によって示されたように、励振トランジスタ対の活性化がＮＯ ＣＵＲＲＥＮＴ（電流なし）条件（すなわち、出力抵抗器Ｒ_outを通して流れる電流が零である）を生じたかどうか判定する。もし励振トランジスタ対の活性化がＮＯ ＣＵＲＲＥＮＴ条件を生じるならば、判定ブロック２３２５から出る“ＹＥＳ”経路に従って、制御論理２２０５は、ブロック２３３０によって示されたように、適当な状態フラグをセットすることになる。
次のステップは、判定ブロック２３３５によって示されたように、前述の励振トランジスタ対の全て６つが活性化されているかどうか判定することである。もし全て６つの励振トランジスタ対が活性化されていないならば、判定ブロック２３３５から出る“ＮＯ”経路に従って、制御論理２２０５は、次の所望励振トランジスタ対を活性化するように異なるホール符号を発生することになる。判定ブロック２３３５から出る“ＹＥＳ”経路に従って、いったん全て６つの励振トランジスタ対が活性化されたならば、状態フラグが下に説明するように故障条件を分離するために必要な情報の全てを含む。
全て６つの励振トランジスタ対が活性化された後、状態フラグは、判定ブロック２３４０によって示されたように、励振トランジスタ対の２つが活性化されたとき増幅器ＳＨＵＴＤＯＷＮが起こったこと、及びこれら２つの影響された励振トランジスタ対が同じ上側又は同じ下側励振トランジスタを含んでいたことを表示することがある。例えば、第１トランジスタ対は励振トランジスタ対Ｔ１及びＴ５を含むトランジスタ対であるかもしれないのに対して、第２トランジスタ対は第２トランジスタ対は励振トランジスタ対Ｔ１及びＴ６を含むトランジスタ対であるかもしれず、ここで両励振トランジスタ対が位相励振器Ａからの上側励振トランジスタＴ１を含む。もし状態フラグがこの条件が起こっていることを表示するならば、判定ブロック２３４０から出る“ＹＥＳ”経路に従って、ブロック２３４５に示されたように、故障条件は増幅器短絡と識別されることになる。特に、故障分離プロセスは、故障条件を引き起こす励振トランジスタ、例えば、上の例における励振トランジスタＴ１を識別することになる。
しかしながら、もし状態フラグがこの状態が起こったことを表示しないならば、判定ブロック２３４０から出る“ＮＯ”経路に従って、状態フラグはその代わりに、判定ブロック２３５０によって示されたように、２つの励振トランジスタ対が活性化されたときＳＨＵＴＤＯＷＮが起こったこと、及び両方の場合に同じ２つの位相励振器が係わりを持たされたことを表示することがある。例えば、状態フラグは、励振トランジスタＴ１及びＴ５を含むトランジスタ対の活性化中ＳＨＵＴＤＯＷＮが起こったこと、及び励振トランジスタＴ２及びＴ４を含むトランジスタ対の活性化中ＳＨＵＴＤＯＷＮが起こったことを表示することがある。この例では、位相励振器Ａ及びＢのみが影響される。もし状態フラグがこの状態が起こっていることを表示するならば、判定ブロック２３５０から出る“ＹＥＳ”経路に従って、ブロック２３５５によって示されたように、故障はモータ巻線短絡と識別されることになる。特に、故障分離プロセスは、上の例では、モータ巻線Ｍ_A−Ｍ_Bのような、故障条件を引き起こすモータ巻線を識別することになる。
もし状態フラグが判定ブロック２３５０内に識別された条件が存在することを表示しないならば、判定ブロック２３５０から出る“ＮＯ”経路に従って、判定ブロック２３６０に示されたように、状態フラグは４つの励振トランジスタ対が活性化されたときＳＨＵＴＤＷＯＮが起こったことを表示することがあり、かつここで全て４つの影響された励振トランジスタ対が共通励振器を伴う。例えば、状態フラグは、励振トランジスタＴ１とＴ５、Ｔ１とＴ６、Ｔ２とＴ４、Ｔ３とＴ４を含むトランジスタ対の活性化中ＳＨＵＴＤＯＷＮが起こったことを表示することがある。この例では、共に位相励振器Ａに係わる上側励振トランジスタＴ１又は下側励振トランジスタＴ４が係わりを持たせられる。もし状態フラグがこの条件が起こったことを表示するならば、判定ブロック２３６０から出る“ＹＥＳ”経路に従って、ブロック２３６５によって示されたように、故障条件はモータ巻線地絡と識別されることになる。特に、故障分離プロセスは、上の例における位相励振器Ａのような、故障条件を引き起こす位相励振器を識別することになる。
もし状態フラグが判定ブロック２３６０内に識別された条件が存在することを表示しないならば、判定ブロック２３６０から“ＮＯ”経路に従って、判定ブロック２３７０に示されたように、状態フラグは２つの励振トランジスタ対の活性化中ＮＯ ＣＵＲＲＥＮＴ条件が起こったことを表示することがあり、かつここで２つの影響された励振トランジスタ対が同じ上側又は下側励振トランジスタに係わる。例えば、状態フラグは、励振トランジスタＴ１及びＴ５を含むトランジスタ対の活性化中かつ励振トランジスタＴ１及びＴ６を含むトランジスタ対の活性化中ＮＯ ＣＵＲＲＥＮＴが起こったことを表示することがある。この例では、影響された励振トランジスタ対の両方が上側励振トランジスタＴ１に係わる。もし状態フラグがこの条件が起こったことを表示するならば、判定ブロック２３７０からから出る“ＹＥＳ”経路に従って、ブロック２３７５によって示されたように、故障条件は増幅器開絡と識別されることになる。特に、故障分離プロセスは、上の例における励振トランジスタＴ１のような、故障条件を引き起こす励振トランジスタを識別することができる。
もし状態フラグが判定ブロック２３７０内に識別された条件が存在することを表示しないならば、判定ブロック２３７０から出る“ＮＯ”経路に従って、状態フラグは４つの励振トランジスタ対の活性化中ＮＯ ＣＵＲＲＥＮＴ条件が起こったことを表示することがあり、かつここで全て４つの励振トランジスタ対が共通位相励振器に係わる。例えば、状態フラグは、励振トランジスタ対Ｔ１とＴ５、Ｔ１とＴ６、Ｔ２とＴ４、Ｔ３とＴ４の活性化中ＮＯ ＣＵＲＲＥＮＴ条件が起こったことを表示することがある。この例では、位相励振器Ａが全て４つの励振トランジスタ対に共通である。もし状態フラグがこの条件が起こったことを表示するならば、判定ブロック２３８０から出る“ＹＥＳ”経路に従って、ブロック２３８５によって示されたように、故障条件はモータ巻線開路と識別されることになる。特に、故障分離プロセスは、上の例におけるモータ巻線Ｍ_Aのような、故障条件を引き起こしつつあるモータ巻線を識別することができる。
もし状態フラグが判定ブロック２３８０内に識別された条件が存在することを表示しないならば、判定ブロック２３８０から出る“ＮＯ”経路に従って、ブロック２３９０によって示されたように、故障分離プロセスは単一点故障条件が分離されていないことを表示することになる。

表ＩＩは、本発明の好適実施例による、三相モータ、ＰＷＭ増幅器故障分離プロセスをまとめたものであり、ここで“ＳＤ”はＳＨＵＴＤＯＷＮを意味し、“ＮＣ”はＮＯ ＣＵＲＲＥＮＴ条件を意味する。重要なことは、三相モータ、ＰＷＭ増幅器サーボ・システム回路２２００が“ＷＹＥ”構成で結線されることである。しかしながら、当業者が容易に理解するように、同様の故障分離プロセスを“ＤＥＬＴＡ（デルタ）”結線された、三相モータ、ＰＷＭ増幅器サーボ制御回路用に実現することができ、同じく本発明の精神に含まれると考えられる。
図２４は、単相モータ、直線増幅器サーボ・システム回路２４００の回路図である。通常動作条件下で、単相モータ、直線増幅器サーボ・システム回路２４００は単相巻線を通る電流の整流を制御し、この電流が単相モータＭ_Sの軸の回転を制御する。
本発明の好適実施例では、単相モータ、直線増幅器サーボ・システム回路２４００は、モノリシック二電力演算増幅器を含む単一供給、直線相互コンダクタンス・ブリッジ増幅器構成を採用する。特に、この構成は電源Ｖ_Lを含み、ここにＶ_Lは好適実施例では＋２８ボルトに等しい。電源Ｖ_Lは双方向負荷流Ｉ_LOADを単相モータＭ_Sの巻線を通して流れさせ、ここで負荷電流Ｉ_LOADは、演算増幅器Ａと演算増幅器Ｂの入力に印加された電圧の量に直線的に比例する。
単相モータ、直線増幅器サーボ・システム回路２４００は、次のように働く。増幅器Ａの出力電圧と電源電圧とが等しく（例えば、＋２８ボルトに等しく）かつ増幅器Ｂの出力電圧が０ボルトに等しいとき、負荷電流Ｉ_LOADはその最大正値にある。増幅器Ｂの出力電圧と電源電圧とが等しくかつ増幅器Ａの出力電圧が０ボルトに等しいとき、負荷電流Ｉ_LOADはその最大負値にある。増幅器Ａと増幅器Ｂの出力電圧とが等しいとき（例えば、増幅器Ａと増幅器Ｂの出力が共にＶ_L／２ボルトに等しいとき）負荷電流Ｉ_LOADは零であることが、論理的に導かれる。更に、単相モータＭ_Sに掛かる電圧は０ボルトと＋Ｖ_Lボルトとの間で変動することができ、したがって、負荷電流Ｉ_LOADがこの電圧に対して直線的に変動する。それゆえ、当業者が容易に理解するように、単相モータ、直線増幅器サーボ・システム回路２４００は、電圧命令の大きさ（すなわち、演算増幅器Ａと演算増幅器Ｂの入力に印加される電圧の量）を変動させることによってサーボ・モータＭ_Sの軸に印加されるトルクの量を直線的に変動させることができる。
図２４で直線増幅器Ａ及びＢの各々は、上側励振トランジスタ（図示されていない）及び下側励振トランジスタ（図示されていない）を含み、これらは図２０に示された上側及び下側励振トランジスタに極く類似している。したがって、直線増幅器Ａ内の上側励振トランジスタは単相モータＭ_Sを通る正負荷電流Ｉ_LOADを発生する任を主として負うのに対して、直線増幅器Ｂ内の下側励振トランジスタは正負荷電流Ｉ_LOADをシンクする。同様に、直線増幅器Ｂ内の上側励振トランジスタは単相モータＭ_Sに対する負負荷電流Ｉ_LOADを発生する任を主として負うのに対して、直線増幅器Ａ内の下側励振トランジスタは負負荷電流Ｉ_LOADをシンクする。
ＰＷＭ増幅器サーボ組立体制御回路２０００及び２２００と異なり、単相モータ、直線増幅器サーボ・システム回路２４００は、電源電流Ｉ_Sの量（すなわち、演算増幅器Ａ及びＢに印加される電流の量）に比例する電圧Ｖ_Sを発生するアナログ電源電流センシング増幅器２４１０、復帰電流±Ｉ_Rの量（すなわち、＋Ｉ_R又は−Ｉ_Rの量）に比例する電圧Ｖ_Rを発生するアナログ復帰電流センシング増幅器２４１５、及び単相サーボ・モータＭ_Sの巻線と並列に接続された分流抵抗器を含み、ここで分流抵抗器は比較的大きなインピーダンス値を有する。これらの追加構成要素は、次の単一点故障条件、すなわち、モータ巻線開路、モータ巻線線絡、モータ巻線地絡、増幅器励振トランジスタ開路、増幅器短絡、及び増幅器出力ワイヤ・ボンド開路を検出しかつ分離することを可能にする。
図２５は、単相モータ、直線増幅器サーボ故障分離プロセスの詳細ステップを示す流れ図である。図２５によれば、制御回路２４０５は、ブロック２５１０によって示されたように、正電圧の所定量を発生することによってその手順を開始する。この電圧は、加算器２４２０及び加算器２４２５に印加される。好適実施例では、発生された電圧の量はＶ_L／ボルト（すなわち、加算器２４２０及び２４２５用規準電圧）より小さく、それであるから増幅器Ａの入力電圧は＋Ｖ_Lボルトより小さく及び増幅器Ｂの入力電圧は＋Ｖ_Lボルトよりかなり小さい。通常動作条件下で、これは、単相モータ巻線を通して、図２４に示されたように、正負荷電流Ｉ_LOADを生じる。次に、ブロック２５１５によって示されたように、復帰電流＋Ｉ_Rが測定される。
次いで、制御回路２４０５は、ブロック２５２０によって示されたように、負電圧の所定量を発生する。この負電圧は、加算器２４２０及び加算器２４２５に同じように印加される。増幅器Ｂはいまや＋Ｖ_Lボルトよりちょっと低い入力電圧を受け及び増幅器Ａは＋Ｖ_Lよりかなり低い入力電圧を受ける。通常動作条件下で、これは、単相モータ巻線を通して負負荷電流Ｉ_LOADを生じることになる。ブロック２５２５に示されたように、復帰電流−Ｉ_Rがいまや測定される。
次いで、制御回路２４０５は、判定ブロック２５３０によって示されたように、復帰電流＋Ｉ_Rの大きさ及び復帰電流−Ｉ_Rの大きさが共に復帰電流の所定（すなわち、期待された）量より小さいかどうか判定し、ここで復帰電流の期待された量Ｉ_Expは故障条件が存在しないとき期待する復帰電流の量に等しい。もし＋Ｉ_R及び−Ｉ_Rの大きさが共に復帰電流の期待された量Ｉ_EXPの大きさより小さいならば、判定ブロック２５３０から出る“ＹＥＳ”経路に従って、ブロック２５３５に示されたように、故障条件はモータ巻線開路と識別されることになる。モータ巻線開路条件はモータ巻線を通して通常流れる電流を、代わりにモータＭ_Sの抵抗負荷に比較して大きな抵抗（すなわち、好適実施例では１４００オーム）を有する分流抵抗器２４３０を通して強制的に流すので、復帰電流＋Ｉ_Rの大きさ及び復帰電流−Ｉ_Rの大きさはＩ_EXPより小さい。したがって、分流電流Ｉ_SHUNTは分流抵抗器２４３０を横断して大きな電圧降下を引き起こす。分流抵抗器２４３０を横断する大きな電圧降下は、そうでなければアナログ電流センシング増幅器２４１５の入力を横断して起こっているであろう電圧降下を減少させる。これが、復帰電流＋Ｉ_Rの大きさ及び復帰電流−Ｉ_Rの大きさが共に復帰電流の期待された量Ｉ_Eの大きさより小さいと云うように、復帰電流＋Ｉ_Rの大きさ及び同様に復帰電流−Ｉ_Rのそれを小さくする。
しかしながら、もし復帰電流＋Ｉ_Rの大きさ及び復帰電流−Ｉ_Rの大きさが期待された復帰電流Ｉ_EXPの大きさ以上であるならば、判定ブロック２５３０から出る“ＮＯ”経路に従って、制御回路２４０５は、判定ブロック２５４０に示されたように、復帰電流＋Ｉ_Rの大きさ及び復帰電流−Ｉ_Rの大きさが共にＩ_EXPに等しいかどうか判定する一方、またモータ軸の運動があるかどうか、又はモータ軸の誤り運動があるかどうか判定することになる。もしこれらの条件が真実ならば、判定ブロック２５４５から出る“ＹＥＳ”経路に従って、ブロック２５４５に示されたように、故障条件はモータ巻線線絡と識別されることになる。
しかしながら、もし復帰電流＋Ｉ_Rの大きさ及び復帰電流−Ｉ_Rの大きさがＩ_EXPに等しくなく、モータ軸運動があり、及び／又はモータ軸の誤り運動があるならば、判定ブロック２５４０から出る“ＮＱ”経路に従って、制御回路２４０５は、判定ブロック２５５０によって示されたように、復帰電流＋Ｉ_Rの大きさ及び復帰電流−Ｉ_Rの大きさが共に零であるかどうか判定する一方、また電源電流Ｉ_Sが復帰電流の量に等しくないかどうか（すなわち、Ｉ_Sが零でないかどうか）判定することになる。もしこれらの条件が真実ならば、判定ブロック２５５０から出る“ＹＥＳ”経路に従って、ブロック２５５５によって示されたように、故障条件はモータ巻線地絡と識別されることになる。
しかしながら、もし復帰電流＋Ｉ_Rの大きさ及び復帰電流−Ｉ_Rの大きさが零に等しくなく及び／又は電源電流Ｉ_Sの大きさが零に等しいならば、判定ブロック２５５０か出る“ＮＯ”経路に従って、制御回路２４０５は、判定ブロック２５６０によって示されたように、復帰電流＋Ｉ_Rの大きさがＩ_EXPの大きさに等しいかどうか、一方復帰電流−Ｉ_Rの大きさが零に等しいかどうか判定するか、又は復帰電流−Ｉ_Rの大きさがＩ_EXPの大きさに等しいかどうか、一方復帰電流＋Ｉ_Rの大きさが零に等しいかどうか判定することになる。もしこれらの条件のどちらかが真実ならば、判定ブロック２５６０から出る“ＹＥＳ”経路に従って、ブロック２５６５によって示されたように、故障条件は増幅器励振トランジスタ開路と識別されることになる。
しかしながら、もし復帰電流＋Ｉ_Rの大きさがＩ_EXPの大きさに等しくない一方復帰電流−Ｉ_Rの大きさが零に等しい、又は復帰電流−Ｉ_Rの大きさがＩ_EXPの大きさに等しくない一方復帰電流＋Ｉ_Rの大きさが零に等しいならば、制御回路２４０５は、ブロック２５７０によって示されたように、復帰電流＋Ｉ_Rの大きさが期待された復帰電流量Ｉ_EXPに等しいかどうか、一方復帰電流−Ｉ_Rの大きさが復帰電流の所定最大量Ｉ_MAXより大きいかどうか判定するか、又は復帰電流−Ｉ_Rの大きさがＩ_EXPに等しいかどうか、一方復帰電流＋Ｉ_Rの大きさが復帰電流の所定最大量Ｉ_MAXより大きいかかどうか判定することになる。もしこれらの条件のどちらかが真実ならば、判定ブロック２５７０から出る“ＹＥＳ”経路に従って、ブロック２５７５によって示されたように、故障条件は増幅器励振トランジスタ短絡と識別されることになる。更に、制御回路２４０５は、保護増幅器ＳＨＵＴＤＯＷＮ命令を発する。
しかしながら、もし復帰電流＋Ｉ_Rの大きさがＩ_EXPに等しくなく及び／又は復帰電流−Ｉ_Rの大きさがＩ_MAXより大きいならば、又はもし復帰電流−Ｉ_Rの大きさがＩ_EXPに等しくなく及び／又は復帰電流＋Ｉ_Rの大きさがＩ_MAXより大きいならば、判定ブロック２５７０から出る“ＮＯ”経路に従って、制御回路２４０５は、判定ブロック２５８０によって示されたように、復帰電流＋Ｉ_Rの大きさ及び復帰電流−Ｉ_Rの大きさが共に零に等しいかどうか、かつ電源電流Ｉ_Sがまた零であるどうか判定することになる。もしこれらの条件が真実ならば、判定ブロック２５８０から出る“ＹＥＳ”経路に従って、ブロック２５８５によって示されたように、故障条件は増幅器出力ワイヤ・ボンド開路と識別されることになる。
最後に、もし復帰電流＋Ｉ_R及び−Ｉ_Rが零でない及び／又は電源電流Ｉ_Sが零に等しくないならば、判定ブロック２５８０から出る“ＮＯ”経路に従って、判定ブロック２５９０によって示されたように、単一故障条件は識別されないことになる。重要なことは、ステップ２５３０から２５８５を実行する順序を変える類似の手順も本発明の精神内に含まれると考えられることである。
表ＩＩＩは、単相モータ、直線増幅器サーボ故障分離能力の故障分離機能をまめたものである。

本発明の好適実施例では、３つの上に識別されたサーボ・システム回路、すなわち、二相モータ、ＰＷＭ増幅器サーボ・システム回路２０００、三相モータ、ＰＷＭ増幅器サーボ・システム回路２２００、及び単相モータ、直線増幅器サーボ・システム回路２４００は、相互コンダクタンス装置として説明される。相互コンダクタンス装置は、電流制御装置であり、この装置では負荷電流（すなわち、モータ巻線を通して流れる電流）が入力電圧命令を制御する負帰還信号として使用される。しかしながら、当業者が容易に理解するように、これら３つのサーボ・システム回路を電流制御装置よりはむしろ電圧制御装置として実現することもでき、そのようにしても上に説明した相当する故障分離プロセスには影響しないであろう。
信号処理
この発明は、ＦＬＩＲ像と、像のディスプレイと目の間のインターフェースの質を高めるよう設計された多くの信号処理技術を提供する。ＦＬＩＲ像とその像の表示の質を高めることにより、この発明はＡＯＩとそのＡＯＩ内のターゲットをより正確に描き、またこれをより安全な離れた範囲で描くことができる。図２６に示すように、この信号処理技術は、２Ｄコントラストフィルタ２６０５、双線形内挿過程（ＢＬＩ）２６１０、ダイナミックレンジ制御フィルタ２６１５、副画素ディザリング過程２６２０、上に述べたＦＬＩＲからレーザへのボアサイト過程中に用いられるＦＬＩＲ集束技術２６２５、などの多数の像処理機能を含む。更に、この信号処理技術は、ノッチフィルタ機能２６３０、アナログ・ディジタル変換機能２６３５、画素または検出器要素値の利得およびレベル訂正機能２６４０、デッドセル置換機能２６４５、アナログ・ディジタル変換器（ＡＤＣ）オフセットパターン除去機能２６５０、などの多数の像前処理機能を含む。
像前処理機能から始めると、ノッチフィルタ２６０５は反射防止膜を有するガラスで製作した光学フィルタである。ノッチフィルタ２６０５は、ＩＲ信号がＦＰＡ２６０７上に集束する前にＩＲ信号から雑音信号を除去するよう設計される。具体的に述べると、ノッチフィルタ２６０５は大気放射により生じる中間波領域（すなわち、４．２乃至５．５５マイクロメートル）内の雑音信号を除去するよう設計される。当業者は周波数のこの領域を大気吸収帯またはＣＯ₂吸収帯と呼ぶことが多い。ノッチフィルタ２６０５がないと、これらの大気雑音信号はＩＲ信号を劣化させてＩＲ像の質を低下させる。
ノッチフィルタ２６０５はＦＬＩＲ検出器／クーラー組立体４２７内に設けられる。ノッチフィルタは一般にこの技術で知られており、一般にＦＰＡコールドフィルタと共に製作される。
ＩＲ像がＦＰＡ２６０７上に集束すると、各検出器要素のアナログ値（すなわち光学像）はディジタル化される。以後、各検出器要素値を画素または画素値と呼ぶ。各検出器要素のディジタル化は４個のアナログ・ディジタル変換器（ＡＤＳ）２６３５の中の１個で行う。しかし当業者が理解するように、３個以下または５個以上のＡＤＣを用いてもよい。この発明の好ましい実施の形態では、各検出器要素は１２ビットの画素値に変換される。
像データのフレーム毎に、１２ビットの画素値に多くの像前処理を行う。第１像前処理技術は利得およびオフセット訂正過程２６４０である。利得およびオフセット訂正過程２６４０の目的は、各画素を校正することにより光学像から特定の雑音成分を除去することである。各画素を校正することにより除去される雑音成分は、検出器要素毎に利得とオフセットが変動することによって生じる。利得とオフセットの変動は、上に述べたディジタル化過程で対応する画素値に伝わる。校正を行うには、各検出器要素にホット基準とコールド基準を適用し、また必要があれば各画素の利得係数とオフセット係数を調整して各画素がホット基準に応じてまたコールド基準に応じて同じ値を反射するようにする。ホット基準とコールド基準に応じて各画素を校正する過程はこの技術でよく知られている。
次の像前処理技術は「デッド」セル置換過程２６４５である。この過程の目的は「デッド」セル（すなわち、正しく応答しない検出器要素）のリストを保持して、各「デッド」セルに対応する画素値を最良の近似値に置換することである。最良の近似値は「デッド」セルに対応する画素に接する画素の値を平均することにより得られる。最良の近似値を得るには、正しく機能している検出器要素に対応する近くの画素だけを用いる。
信号処理サブシステムはよく知られている基準のどれかを適用して、どの検出器要素が「デッド」であるか判定する。例えば、各検出器要素の熱応答と期待応答とを比較する。実際の応答が期待応答より非常に大きいか非常に小さい場合は、対応する検出器要素は恐らく正しく機能していない。検出器要素が正しく機能していないと判定するのによく用いられる別の基準は、検出器要素のディジタル応答が安定しているか、または変動しているように見えるかである。応答が変動している、すなわち揺らいでいるのは、恐らく対応する検出器要素が正しく機能していないことを示す。更に別の基準は、所定の検出器要素の応答と、全ての検出器要素の応答から得た平均値とを比較することである。或応答が平均の応答と実質的に異なる場合は、これは恐らく対応する検出器要素が正しく機能していないことを示す。また所定の検出器要素のダイナミックレンジが限定されている場合は、これは恐らく検出器要素が正しく機能していないことを示す。当業者が理解するように、この基準のリストはこれだけではなく、他の基準を同様に用いて「デッド」検出器要素を識別することができる。一般に、「デッド」セルを置換する手続きはこの技術でよく知られている。
次の信号前処理技術はＡＤＣオフセットパターン除去機能２６５０である。ＦＰＡ２６０７は、図２６に示すように４本の出力線を有する。これらの各出力線は別個のＡＤＣ２６３５に接続する。上に説明したように、ＡＤＣは検出器要素に関連するアナログ電圧レベルを１２ビットの画素値に変換する。更に、４個の各ＡＤＣはＦＰＡ３６０７の４列毎の電圧レベルをそれぞれ変換する。例えば、第１ＡＤＣは列１、５、９、．．．、４７７内の検出器要素に関連するアナログ電圧を変換する。第２ＡＤＣは列２、６、１０、．．．、４７８内の検出器要素に関連するアナログ電圧を変換する。第３ＡＤＣは列３、７、１１、．．．、４７９内の検出器要素に関連するアナログ電圧を変換する。第４ＡＤＣは列４、８、１２、．．．、４８０内の検出器要素に関連するアナログ電圧を変換する。更に、ＡＤＣは非常に感度が高く、特に大気温度が変化すると、時間と共にドリフトする傾向がある。しかしＡＤＣのどれかがドリフトする場合は、他の３個のＡＤＣとは無関係にドリフトする可能性がある。１個のＡＤＣが他の３個に対してドリフトすると、ディジタル像の４列毎に望ましくないオフセットすなわちバイアスを生じる。このオフセットすなわちバイアスをＡＤＣオフセットパターンと呼ぶ。
ＡＤＣオフセットパターン除去機能２６５０の目的は、影響を受けた画素値を調整することによりディジタル像からこのオフセットすなわちバイアスを除去することである。この発明の好ましい実施の形態では、影響を受けた画素値を調整するこの過程は次のように行う。像データのフレーム毎に、４つのヒストグラム２６６０、２６６５、２６７０、２６７５を生成する。各ヒストグラムの内容は、４個のＡＤＣの中の対応するものが作った画素値に基づく。したがって、各ヒストグラムは１２０列の画素値を反映し、各列は４８０画素値を含む。像データのフレーム毎に、信号処理ソフトウエア２６８０は１つのヒストグラムについて平均画素値Ｈ₁、Ｈ₂、Ｈ₃、Ｈ₄を順に計算する。ここで、２０番目の百分順位の画素値と８０番目の百分順位の画素値の間の画素値だけを用いて平均画素値を計算する。また像データのフレーム毎に、信号処理ソフトウエアは、４個の個別のヒストグラム平均値Ｈ₁、Ｈ₂、Ｈ₃、Ｈ₄を全て用いて、全画素値平均Ｈ_BARを計算する。Ｈ_BARと個別の各ヒストグラムの平均値との差Ｈ_BAR−Ｈ₁、Ｈ_BAR−Ｈ₂、Ｈ_BAR−Ｈ₃、Ｈ_BAR−Ｈ₄を、場合に応じて、既存の対応する画素オフセット係数に加えるかまたは画素オフセット係数から引く。
前に述べたように、この発明ではディジタル像の質とディスプレイと目の間のインターフェースを改善するために多くの像処理機能を用いる。これらの像処理機能の第１は２Ｄ鮮鋭化フィルタ２６０５である。２Ｄ鮮鋭化フィルタ２６０５はエッジの強化（すなわち、高周波の像データの強化）に用いる。一般にエッジの強化を行うには、画素入力像当たり１２ビットに低域フィルタリングを行って低域像を生成する。入力像から低域像を引くと高域像が得られる。次に、低域像と高域像の相対利得を調整した後、２つの像を統合して強化された像を形成する。
図２７は２Ｄ鮮鋭化フィルタ２７００の好ましい実施の形態であって、次のように動作する。２Ｄ鮮鋭化フィルタ２７００は、４８０ｘ４８０画素の入力像内の各画素に低域フィルタリング操作２７０５を行って低域像を生成する。低域像は低周波の像データを含むのでやや不鮮明に見える。低域フィルタリング操作は３ｘ３たたみこみ過程であって、入力像内の各画素の値を平均画素値で置換する。入力像内の任意の所定の画素について、画素値と各近接画素値を合計し、次に合計を得るのに用いた画素の数でその合計を割って、その画素の平均画素値を計算する。入力像の外のエッジ上にない全ての画素では、各合計操作に９画素が関係する。すなわち、平均化操作を行っている１画素と、それに近接する８画素である。この平均化操作を、入力像内の全ての画素値について繰り返す。
２Ｄ鮮鋭化フィルタ２７００はまた、入力像内の各画素値から低域像内の各画素値を引いて高域像を生成する。引き算を加算器２７１０で表す。この引き算の結果得られる高域像は入力像からの高周波の像データを含む。
更に２Ｄ鮮鋭化フィルタ２７００は、像コントラスト測度２７１５を生成する。像コントラスト測度２７１５を生成するには、まず入力像の各行に沿って隣接する画素値の差を計算する。次に全ての差の値を合計することによりコントラスト測度２７１５が得られる。例えば、真っ白または真っ黒の入力像（すなわち、殆ど低周波成分だけを含む入力像）は、非常に低いコントラスト測度を生じる。当然、雑音成分（一般に高周波雑音）は常にいくらか存在する。しかし、１つおきの画素が交互に白と黒である（すなわち、入力像が大きな高周波成分を含む）チェッカ盤パターンを示す入力像は、非常に大きなコントラスト測度を生じる。
次に２Ｄ鮮鋭化フィルタ２７００は、低域像２７２０の利得レベルＧ_Lを高域像２７２５の利得レベルＧ_Hに対して、またはその逆に、調整する。低いコントラスト測度を有する入力像は一般に比較的低い信号対雑音比（ＳＮＲ）を示す。高周波の雑音内で不鮮明になった像の詳細を強化するには、比Ｇ_L／Ｇ_Hを高くする。これは、信号の利得レベルを上げ、雑音の利得レベルを下げる効果を持つ。次に調整された利得レベルＧ_LとＧ_Hを低域像と高域像内の各画素にそれぞれ与える。高いコントラスト測度を有する入力像は一般に高いＳＮＲを示す。像の質を更に高めるには比Ｇ_L／Ｇ_Hを低くする。これは、入力像内にすでに存在する高周波信号を更に強化する効果を持つ。再び、調整された利得レベルＧ_LとＧ_Hを低域像と高城像内の各画素にそれぞれ与える。非常に高いコントラスト測度限界と非常に低いコントラスト測度限界の間のコントラスト測度を有する入力像については、コントラスト測度と利得レベルの関係を確立する多項曲線に基づいて比Ｇ_L／Ｇ_Hの調整を行う。この発明の好ましい実施の形態では、比Ｇ_L／Ｇ_Hを調整するのに用いる多項曲線はルックアップテーブルにより実現する。しかし当業者が容易に理解するように、多項曲線は式で容易に実現することができる。
次に２Ｄ鮮鋭化フィルタは、今調整された低域像内の各画素と今調整された高域像内の対応する画素を加えて、強化された像を生成する。加算を加算器２７３０で表す。
この発明で用いられる第２像処理機能は双線形内挿である。双線形内挿は、水平または垂直に像を移すのに用いられる。また像を回転したり電子ズームを与えたりするのにも用いられる。双線形内挿は一般にこの技術でよく知られている。
この発明で用いられる第３像処理機能はダイナミックレンジ制御機能２６１５である。ＬＡＮＴＩＲＮなどの従来のシステムでは、ＩＲ像を強化するのに適応非線形マッピングが用いられている。非線形マッピング方式は目盛りの中央を中心とするガウス分布かフラット分布を用いる。しかしすでに指摘されているように、多くの「見やすい」強さ分布は、人の目がグレイレベルの変化に対して敏感な黒の方を白より強調する。したがって目盛り中央を中心とするフラット分布やガウス分布は、一般に最も「見やすい」像は作らない。
対照的に、この発明は画素像データ当たり１２ビットを扱うので、図２８に示すように、ヒストグラム分布の中心は目盛りの中央ではない。この発明はレイリー分布を用いる。その中心はヒストグラムダイナミックレンジの一端寄りに移動して暗い強さを強調し、明るい強さは強調しない（すなわち、飽和を減少させるため）。
次にこの発明は画素ＩＲ像当たり１２ビット（４０９６個の量子化された値）を画素像当たり８ビット（２５６個の量子化された値）に再マップして、標準の、画素当たり８ビットのＲＳ−１７０ビデオディスプレイ装置に適応させる。再マッピング過程で解像度が実質的に減少するのを避けるために、この発明はルックアップテーブルを用いて１２ビットの像を不均一に８ビットの像に再マップする。図２９に示すように、この発明は１２ビットの像分布の或領域にわたって、特に像データの濃度が高いダイナミックレンジの部分で、１：１マッピング方式と高レベルの解像度を保持する。同時にこの発明は、他の、像データの濃度が低いダイナミックレンジの部分の解像度レベルを減少させる。これにより、画素ディスプレイ装置により８ビットという制限があるにも関わらず、全体の像の解像度を最大にすることができる。
更に第４像処理機能は副画素ディザリング機能２６２０である。この発明では、ＦＰＡ２６３０は５１２ｘ５１２のステアリングアレイである。しかし、標準ＲＳ−１７０ディスプレイ装置は像データの４８０行だけしか走査することができないので、ＩＲ像を生成するにはステアリングアレイの４０８ｘ４８０部分だけが用いられる。いずれにしても、ＦＰＡ２６３０は、アレイの更に小さい部分でも読み出すことのできるウインドウイングモードを有する。例えば、４８０ｘ４８０像を読み出すのに必要な時間の約１／４で２４０ｘ２４０像を読み出すことができる。この発明は、ＦＰＡウインドウイング機能とＦＳＭ４１５を用いて、２倍(2X)に強化された像モードを与える。より具体的に述べると、ＲＳ−１７０ビデオを作成するのに、この発明はＦＳＭ４１５を用いて、直径方向および下方に検出器の中心間の距離の１／２だけＦＰＡのＬＯＳをディザし、２４０ｘ２４０像に基づいて、強化された４８０ｘ４８０の像を得る。
像データを組み合わせてＲＳ−１７０ビデオデータを作る従来の方法では、一般に解像度が減少する。例えば、図３０に示す従来の方法の１つは、偶数番目の検出器の行を捨てて「奇数」ビデオフィールドを作り、「奇数」番目の検出器の行を捨てて「偶数」ビデオフィールドを作る。この方法では、像データの半分を捨てるので検出器の感度が減少する。図３１に示す別の従来の方法は、行を平均して「偶数」と「奇数」ビデオフィールドを作る。この方法では、縦の解像度が減少する。
この発明では、ＦＰＡ２６３０は図３２に示すように２４０ｘ２４０像を用い、図３３に示すように、位置１ａ、２ａ、．．．、２４０ａの像を統合する。次に第１ステップ命令をＦＳＭ４１５用の既存のフィードバック制御命令に加える。ステップ命令により、ＦＳＭ４１５はＦＰＡ２６３０のＬＯＳを検出器の中心間の距離の１／２だけ右にディザする。これにより、図３３に示すように位置１ｂ、２ｂ、．．．、２４０ｂの画素値に基づいて新しい２４０ｘ２４０像が得られる。この像を統合して、位置１ｂ、２ｂ、．．．、２４０ｂの画素値と最初の２４０ｘ２４０像からの位置１ａ、２ａ、．．．、２４０ａの画素値を交互に配置して、第１の２４０（垂直）ｘ４８０（水平）の像を生じる。この第１の２４０ｘ４８０の像は「奇数」ビデオフィールドを表す。
第２ステップ命令により、ＦＳＭ４１５はＦＰＡ２６３０のＬＯＳを検出器の中心間の距離の１／２だけ左に、また検出器の中心間の距離の１／２だけ下にディザする。これによりＦＰＡのＬＯＳは、図３３に示すように位置１ｃ、２ｃ、．．．、２４０ｃにある。この像を統合した後、更に別のステップ命令により、ＦＳＭ４１５は再びＦＰＡ２６３０のＬＯＳを検出器の中心間の距離の１／２だけ右にディザする。これによりＦＰＡ２６３０のＬＯＳは、図３３に示すように位置１ｄ、２ｄ、．．．、２４０ｄにある。この像を統合し、位置１ｄ、２ｄ、．．．、２４０ｄの画素値と位置１ｃ、２ｃ、．．．、２４０ｃの画素値を交互に配置して、第２の２４０（垂直）ｘ４８０（水平）像を生じる。この第２の２４０ｘ４８０像は「偶数」ビデオフィールドを表す。次に「奇数」および「偶数」ビデオフィールドを交互に配置して、元の２４０ｘ２４０像の強化された４８０ｘ４８０像を生じる。当業者が容易に理解するように、この強化された像は元の２４０ｘ２４０ウインドウの２倍に強化された像である。
ＦＬＩＲ集束過程などの他の像処理機能は、上に述べたＦＬＩＲからレーザへのボアサイトなどのサブシステムを支援する。図１１に戻って、ボアサイト焦点板パターン１１００は多数のチェッカ盤パターンを含む。３つの２０８５マイクロラジアンパターンはＷＦＯＶのＦＬＩＲ集束過程に関連し、４つの小さい７００マイクロラジアンパターンはＮＦＯＶのＦＬＩＲ集束過程に関連する。図１２に示すＷＦＯＶおよびＮＦＯＶ集束手続きの間は、ＢＳＭ内のＩＲ源１０１５は対応するチェッカ盤パターンでＦＰＡ２６３０を照射する。重要なことは、ＦＰＡ２６３０上に照射されるチェックの幅は検出器要素の幅より小さく（すなわち、チェッカとチェッカの間の距離は画素と画素の間の距離より小さく）、当業者が容易に理解するように、チェッカ盤パターンと画素の間にランダム位相関係がなければならないことである。ＦＬＩＲ集束レンズを調整するときにチェッカ盤パターンで照射される画素の値に基づいて、多数のコントラスト測定値が得られる。最良の結果を生じるＦＬＩＲ集束レンズの位置は、ピークのコントラスト測定値に対応する。ＦＬＩＲ集束過程について、以下に詳細に説明する。
図３４はＦＬＩＲ集束過程の詳細なステップを示す流れ図である。上に述べたように、ＦＬＩＲ集束過程（すなわち、ＮＦＯＶまたはＷＦＯＶのＦＬＩＲ集束過程）は、ブロック３４０５に示すように、ＩＲ源１０１５が対応するチェッカパターンでＦＰＡ２６３０を照射するときに始まる。次にブロック３４１０に示すように、チェッカ盤パターンで照射された画素の画素値を記録する。次にブロック３４１５に示すように、上に述べた画素値を用いて、最大の記録された画素値と最小の記録された画素値の差を計算することにより、コントラスト測定値を計算する。次にブロック３４２０に示すように、ＦＬＩＲ集束レンズを増分して調整する（すなわち、変換する）。追加のコントラスト測定値が必要な場合は、判定ブロック３４２５からの「はい」の経路に従って追加の画素値を記録し、追加のコントラスト測定値を計算する。しかし追加のコントラスト測定値が必要でない場合は、判定ブロック３４２５からの「いいえ」の経路に従って、ブロック３４３０に示すように調整データをプロットし（すなわち、コントラスト測定値に対するＦＬＩＲ集束レンズ位置を）、ブロック３４３５に示すように、調整データ点を最も適合する多項曲線でつなぐ。次にブロック３４４０に示すように、最大調整点を決定する（すなわち、多項曲線のピーク）。最大調整点は最良のＦＬＩＲ像焦点に対応するＦＬＩＲレンズ位置を表す。
この過程はより明確でより視覚的に正確な像を与えるだけでなく、より正確なボアサイト過程を容易にし、そして最終的に、より正確なＦＬＩＲからレーザへのＬＯＳを与える。更に、上に述べたようにチェッカパターンと画素の間にランダム位相関係がない場合は、記録された画素値はチェッカ盤パターンに対する固定エイリアス（alias）を反映し、ＦＬＩＲ集束レンズを調整しても記録された画素値のコントラストレベルに影響を与えない。
この発明について好ましい実施の形態を説明した。しかし当業者に明らかなように、上に説明したものとは異なる特定の形式で、しかもこの発明の精神から逸れることなく、この発明を実現することができよう。上に説明した好ましい実施の形態は単なる例であって、いかなる意味においても制限するものと考えてはならない。この発明の範囲は、これまでの説明ではなくて特許請求の範囲で与えられるものであり、請求の範囲内に含まれる全ての変更や同等物はその中に包含されるものである。background
The present invention relates to an integrated infrared forward monitoring device / laser sensor. In particular, the present invention relates to a mid-wave infrared forward monitoring (FLIR) subsystem and a laser subsystem including a laser range receiver (laser distance receiver: LRR) and a laser spot tracker (laser spot tracker: LST); Relates to a targeting and imaging system.
The FLIR system utilizes an infrared (IR) detector array that generates an image based on infrared (IR) emission from the (specific) area of interest (AOI). For example, in military applications, the AOI may include targets such as tanks, trucks, and / or other military vehicles and military hardware. Because these targets release heat, they are typically warmer than the surrounding environment. They can therefore be identified in the IR image generated by the FLIR system.
The use of lasers with FLIR systems is well known in the prior art. For example, a laser can be used to indicate a particular target found in the FLIR image. In one conventional FLIR / laser system, laser energy is swept across the target seen in the FLIR image and used to generate a three-dimensional image of the target. The three-dimensional image can then be used for target recognition and / or classification (US Pat. No. 5,345,304). In another conventional FLIR / laser system, a laser is used to determine the distance from the FLIR / laser system host platform to the target (US Pat. No. 4,771,437). In yet another conventional FLIR / laser system, a laser is used to determine the relative position and velocity of the target (US Pat. No. 4,574,191). In addition, lasers are used to direct laser guided munitions to the desired target found in the FLIR image.
In each of the conventional FLIR / laser systems described above, the ability of the FLIR / laser system to accurately recognize, detect, locate and / or track the target depends on the ability of the system to maintain accurate alignment between the FLIR and the laser. Determined. Overflow of the laser occurs due to arbitrarily fixed misalignment between the line of sight of the FLIR (line of sight: LOS) and the LOS of the laser. As shown in FIG. 1, a laser overflow is defined as an unscheduled amount of laser energy 110 that misses the target 105 and is reflected from the background. R in FIG. 2 due to laser overflow _err May cause a ranging error. Second, incorrect distance information will result in more inaccurate target recognition, detection, orientation and velocity information as well as inaccurate weapon guidance data.
Boresighting is a general term in the prior art for the LOS alignment process of a given system. In conventional designs such as low altitude night terrain following infrared navigation (LANTIRN) systems, a boresighting process is utilized to minimize fixed alignment errors between the FLIR LOS and the laser LOS. Typically, the boresighting process typically involves optical and / or mechanical realignment (realignment) of, for example, FLIR LOS and laser LOS. Furthermore, the boresighting process can be manual or automatic. As mentioned above, the boresighting process is generally well known in the prior art.
Unfortunately, for example, alignment errors between FLIR LOS and laser LOS are not necessarily fixed errors. In military applications, FLIR / laser based systems are typically installed on a mobile platform such as a tactical aircraft (eg, F-15 or F-16). These platforms subject the FLIR / laser based system to significant mechanical forces and vibrations. These forces and vibrations act directly on the optical components that govern the FLIR LOS and the laser LOS. Furthermore, the displacement of FLIR LOS and laser LOS about the pitch axis appears to have the most detrimental effect on system performance (eg, the ability to accurately recognize, detect, locate and / or track the target).
As shown in FIG. 3, conventional designs such as LANTIRN utilize separate FLIR optics (optical) pitch bearings 205 and laser optics pitch bearings 210 as well as separate FLIR apertures 215 and laser apertures 2220. Thus, due to the aforementioned mechanical forces and vibrations acting on the FLIR / laser-based system, FLIR LOS and laser LOS are tilted around the pitch axis independently of each other, in addition to any existing alignment errors, LOS jitter and dynamic ( That is, a continuously changing FLIR LOS to laser LOS alignment error occurs. Although the above-described boresighting process can be used to correct the fixed alignment error, it is generally not effective for correcting the dynamic alignment error.
Another problem associated with prior art systems such as LANTIRN, which can contribute significantly to LOS alignment errors, is the fact that the FLIR image rotates around the roll axis as a function of the gimbal pitch angle. To compensate for this unusual property, conventional designs such as LANTIRN reverse rotate the entire FLIR detector assembly. However, the FLIR detector assembly is relatively large and there are many drawbacks to rotating a large mass against a rapidly changing gimbal pitch angle. As a first major drawback, it is very difficult to reverse rotate a large mass with a response time sufficient to compensate for high speed pitch rotation. FLIR LOS vs. laser LOS alignment errors may be added because high speed pitch rotation cannot be compensated. Second, the wires connecting to the FLIR detector array elements must pass through the rotating interface. Rotating the interface and the wires that pass through it has a significant impact on system reliability.
wrap up
The present invention is a high resolution, gimbal intermediate wave FLIR / laser based system that minimizes FLIR LOS versus laser LOS alignment errors, including fixed and dynamic alignment errors, for more accurate target recognition, detection, localization and And / or an electro-optic subsystem designed to provide tracking information. When used with a military weapon launch system, their enhanced performance increases the viability of host platforms that can fire their weapons at longer (safer) distances in the enemy's environment.
In addition, the present invention includes several other subsystems and subsystem capabilities that support and further enhance the effectiveness of the electro-optic system. For example, the present invention includes a “dead” detector cell replacement function, a scene-based pattern removal function, a two-dimensional sharpening filter, a dynamic range control function, and a 2X image enhancement function utilizing a unique subpixel dithering process. A single processing subsystem is included that provides several important new processing and preprocessing functions.
The present invention further includes a new fault isolation subsystem. The fault isolation subsystem can distinguish between fault conditions arising from the amplifier section of various servo systems and fault conditions arising from the servo motor section of the servo system. Accordingly, maintenance personnel need only remove and replace defective portions of the servo system, and do not need to remove and replace the entire servo system.
Finally, the present invention includes a new electromagnetic interference (EMI) grating. This grid more completely prevents unwanted energy from entering the system and interfering with electrical signals. The grid also prevents unwanted energy generated by the system from radiating and interfering with the operation of other nearby systems.
It is an object of the present invention to provide a high resolution, FLIR / laser based targeting and imaging system.
It is another object of the present invention to provide a high resolution, FLIR / laser based system that minimizes alignment errors between FLIR LOS and laser LOS.
It is yet another object of the present invention to minimize alignment errors caused by FLIR LOS and laser LOS jitter by providing a single pitch bearing and a common aperture for FLIR and laser optics.
Alignment errors caused by rotation of the FLIR image around the roll axis when the pitch / yogimbal rotates around the pitch axis can be minimized by reverse rotation of the deroll prism optics instead of the FLIR detector assembly It is yet another object of the present invention.
It is another object of the present invention to filter out unwanted electromagnetic energy from IR energy entering the system aperture.
It is yet another object of the present invention to provide several signal processing functions that further enhance the quality of FLIR images.
Finally, it is an object of the present invention to provide a fault detection process that accurately isolates fault conditions and limits the removal and replacement of hardware that should function properly.
The foregoing and other objects of the present invention include targeting and imaging that includes an infrared forward monitoring (FLIR) optical subsystem that receives infrared (IR) energy from an area of interest (AOI) and generates an IR image of the AOI. Achieved by the system. The system also includes a laser light subsystem that generates laser energy that illuminates at least one object in the AOI and that receives laser energy reflected by the at least one object. In addition, the laser light subsystem and the FLIR light subsystem share a common pitch bearing.
The foregoing and other objects of the present invention also provide an IR image by infrared forward monitoring (FLIR) optics that receives infrared (IR) energy from an area of interest (AOI), and IR energy received from the AOI. Achieved by a targeting and imaging system including a FLIR optical imager that generates The FLIR optical imager is arranged to receive IR energy from the FLIR optical system. The system also directs laser energy from the laser transmitter, laser receiver, and laser transmitter to a desired target located within the AOI and directs laser energy back from the desired target within the AOI to the laser receiver. Laser optics are also included. In addition, FLIR optics and laser optics share a common pitch bearing so that all optical elements that are individually pitch rotated are shared by FLIR optics and laser optics.
The above and other objectives of the present invention also provide for steering the infrared (IR) line of sight (LOS) toward the desired area of interest (AOI), receiving IR energy from the AOI, and reducing IR energy. This is accomplished by a targeting and imaging system that includes infrared forward monitoring (FLIR) optics to focus and generate an AOI optical image. This system includes a laser transmitter, a laser range receiver (LRR), a laser spot tracker (LST), and a laser LOS that steers, receives and receives laser energy so that the transmitted laser energy illuminates at least a portion of the AOI. Also included are laser optics that direct laser energy into the LRR and LST. Further, the FLIR optics means and the laser optics means share a single pitch bearing, and the IR energy and laser energy pass through a common aperture.
The foregoing and other objects of the present invention are also achieved by a targeting and imaging system that includes LOS correction means that adjust IR LOS and laser LOS to minimize LOS alignment errors between IR LOS and laser LOS. Is done.
The foregoing and other objects of the present invention are also achieved by a targeting and imaging system that includes fault isolation means for isolating electrical faults in a servo system including a servo motor and amplifier.
The foregoing and other objects of the present invention are also achieved by a targeting and imaging system that includes a boresight subsystem.
The foregoing and other objects of the present invention are also achieved by a targeting and imaging system that includes a signal processing subsystem.
The foregoing and other objects of the present invention are also achieved by a targeting and imaging system housed in a housing that includes a window through which IR and laser energy passes.
[Brief description of the drawings]
The objects and advantages of the present invention can be understood by reading the following detailed description in conjunction with the following drawings.
FIG. 1 is a diagram showing the concept of laser overflow,
FIG. 2 is a diagram showing the concept of distance measurement error due to laser overflow,
FIG. 3 shows the optical design of a FLIR / laser system according to the prior art,
FIG. 4 is a diagram showing a photoelectric subsystem of a preferred embodiment of the present invention;
FIG. 5A is a diagram showing an optical configuration of a FLIR relay / FOV assembly in NFOV mode;
FIG. 5B is a diagram showing an optical configuration of the FLIR relay / FOV assembly in the WFOV mode;
FIG. 6 is a diagram showing the polarization of the laser energy in the laser compensator unit.
FIG. 7 is a block diagram of the LOS / servo subsystem,
FIG. 8 is a diagram showing the optical configuration of a common pitch / yaw focal (infinite focus) and gimbal mirror in the pitch / yogimbal assembly;
FIG. 9 shows a high speed steering mirror assembly;
FIG. 10 shows the boresight module.
FIG. 11 is a diagram showing a bore sight reticle pattern;
Figure 12 is a block diagram of the boresight process,
FIG. 13 is a diagram showing a part of a bore sight reticle pattern used for aligning laser spots;
14A to 14F are diagrams showing grating patterns used for aligning laser spots;
FIG. 15 is a diagram showing another embodiment of the present invention using an integrated LST / LRR;
FIG. 16 is a diagram showing an integrated LST / LRR;
Figure 17 shows a segmented window;
FIG. 18 shows typical energy artifacts that interfere with FLIR images when an EMI grating according to the prior art is used;
FIG. 19 shows a preferred embodiment of an EMI grating according to the present invention;
FIG. 20 is a diagram of a two-phase motor, PWM amplifier servo system,
FIG. 21 is a flowchart showing a two-phase motor, PWM amplifier failure isolation process,
FIG. 22 is a diagram of a three-phase motor, PWM amplifier servo system,
FIG. 23 is a flow diagram showing a three-phase motor, PWM amplifier failure isolation process,
FIG. 24 is a diagram of a single phase motor, linear amplifier servo system,
FIG. 25 is a flow diagram showing a single phase motor, linear amplifier fault isolation process;
FIG. 26 is a block diagram illustrating signal processing functions according to a preferred embodiment of the present invention;
FIG. 27 shows a two-dimensional sharpening filter;
FIG. 28 is a diagram showing a process of remapping the dynamic range of pixel image data to emphasize dark intensity rather than bright intensity;
FIG. 29 is a diagram showing a non-uniform dynamic range remapping scheme;
FIG. 30 is a diagram illustrating a method for generating even and odd video fields according to the prior art;
FIG. 31 is a diagram illustrating a method for generating even and odd video fields according to the prior art;
FIG. 32 shows a 240 × 240 pixel window in a larger digital image;
FIG. 33 is a diagram illustrating a method for generating even and odd video fields according to a preferred embodiment of the present invention;
FIG. 34 is a flowchart showing a method for forming a FLIR image.
Detailed description
The present invention relates to a targeting and imaging system including an opto-electric subsystem, an EMI grid, a fault isolation subsystem, and a signal processing subsystem. The optoelectric subsystem is a high resolution, gimbal, intermediate wave (3-5 micron) or long wave (8-12 micron subsystem) FLIR; laser range receiver (LRR) for aiming and targeting; and assertive targets Includes a laser spot tracker (LOS) for identification. The FLIR subsystem also provides two optical fields of view (FOV) including 1.2 ° FOV and 3.8 ° FOV with 2X enhancement mode and 4X and 8X electronic zoom modes. Other important features of the present invention include the commonality of optical elements to minimize dynamic alignment errors between FLIR LOS and laser LOS, to minimize bore bending and maximize boresight storage. A separate optical bed and an internal boresight subsystem that minimizes fixed alignment errors between the FLIR LOS and the laser LOS are included.
The terms “optic” and “optical” are typically related to the field of view and field of view. However, hereinafter the terms “optic” and “optical” are more commonly associated with electromagnetic radiation in general and / or devices that are invisible but sensitive to such electromagnetic radiation (ie, IR energy and laser energy). Yes.
Furthermore, the terms FLIR or IR image, light image and digital image will come later. The term IR image relates to the IR energy pattern generated by FLIR optics. The term optical image relates to an analog electronic signal array that together provides an electronic representation of the IR image. The analog electronic signal is generated by an IR detector element array that is responsive to the IR energy pattern of the IR image. The term digital image relates to an array of digital values, also known as pixel values. The pixel value array together provides a digital representation of the light image, with each pixel value associated with a corresponding analog signal value in the light image. Those skilled in the art will readily recognize that light and digital images are not visible images, but are arrays of analog and digital values, respectively. Those skilled in the art will also readily recognize that visible images can be created from light and digital images with appropriate display hardware.
Optoelectric subsystem
As shown in FIG. 4, in the preferred embodiment of the present invention, IR energy enters a segmented target acquisition window (not shown) and is collected by a common pitch / iodine afocal 401. The common pitch / yogimbal afocal 401, or common aperture, is said to be “common” because not only laser energy but also IR energy is transmitted and received using the same aperture. The common aperture 401 is actually a set of lenses including one positive lens 401a and two negative lenses 401b and 401c, as shown in FIG. 5A. This set of lenses equalizes the index of refraction for both IR energy and laser energy, allowing a single aperture to be used for both laser energy and IR energy. By having one aperture for both laser energy and IR energy, both the fixed and dynamic alignment errors between the FLIR LOS and the laser LOS are significantly reduced.
The IR energy beam as it enters the system through the common aperture 401 is reduced in diameter and directed to the rotating mirror 403. Next, the rotating mirror 403 directs the IR energy beam downward toward the pitch axis center line 405. The rotating mirror 403 then redirects the laser energy and IR energy again parallel to the system longitudinal axis where it strikes the dichroic D1 located in the FLIR relay / FOV assembly 407. Dichroic D1 separates the incident energy based on the spectral content. Dichroic D1 does this by propagating energy having a wavelength greater than ˜2.7 microns (eg, IR energy) and reflecting energy having a wavelength shorter than ˜2.0 microns (eg, laser energy). .
The FLIR relay / FOV assembly 407 includes several fixed lenses 411a and 411b and several movable lenses 412a and 412b. As will be described below, these lenses act on IR energy to reduce or increase the IR beam diameter depending on whether the FLIR relay / FOV assembly 407 is in a narrow FOV (NFOV) mode or a wide FOV (WFOV) mode, respectively. To do. The IR energy is then passed to dichroic D2, which propagates IR energy, similar to dichroic D1.
The IR energy then strikes the deroll assembly 413. The deroll assembly 413 includes a deroll prism 414 and a servomotor (not shown) that rotates the deroll prism 414. The deroll prism 414 internally reflects the IR beam an odd number of times, for example, three times, and causes the IR energy to be emitted from the deroll prism 414 in the same direction as the incident direction. When the deroll prism 414 is rotated around the axis defined by the IR ray by the servo motor, there is an effect of reversely rotating the FLIR image at a ratio of 2: 1 (for example, by rotating the deroll prism 414 by + 22.5 °). The IR image rotates by −45 °), which counteracts the rotation of the FLIR image caused by the change in pitch angle as described above. It is called a double effect. The deroll prism 414 will be described later.
Next, a fast steering mirror (FSM) 415 included in the FSM assembly 416 reflects the IR energy beam through the dichroic D3 located in the FLIR imager / focus assembly 417. The FSM 415 is used to eliminate motion-induced image blur associated with excellent LOS stabilization and gimbal scanning and ground rush light flow. The FSM 415 will also be described later.
The FLIR imager / focus assembly 417 includes an imager afocal 419 that translates along a linear bearing to compensate for focus variations due to temperature and altitude variations. Thus, imager afocal 419 can adjust the focus of IR energy over a range of temperatures and altitudes.
The FLIR imager / focus assembly 417 is connected to the FLIR detector / cooler assembly 427. To facilitate FLIR detection and imaging, a lens set including lenses 425a, 425b collimates IR energy as it passes through the FLIR imager / focus assembly, FLIR detector / cooler assembly interface. By collimating the IR energy, the IR image is much less sensitive to small alignment errors between the FLIR imager / focus assembly 417 and the FLIR detector / cooler assembly 427.
As noted above, FLIR relay / FOV assembly 407 provides both NFOV and WFOV for the area of interest. As shown in FIG. 5A, in the NFOV mode, the FLIR relay / FOV assembly 407 utilizes a first lens set 501 and a second lens set 502. It should be understood that the first lens set 501 and the second lens set 502 are the single lenses 411a and 411b in FIG. Both the first lens set 501 and the second lens set 502 represent achromatic doublets (ie, positive and negative lens pairs) and are both fixed with respect to the IR optical path. Individually, the first lens set 501 focuses the IR energy and the second lens set 502 re-collimates the IR energy. Both the first lens set 501 and the second lens set 502 provide a maximum afocal magnification and thus a minimum FOV. It should be appreciated that in the NFOV mode, there are no moving optical elements in the FLIR relay / FOV assembly 407. Moving optical elements are likely to cause alignment errors, and boresight storage and integrity are most important in the NFOV mode.
As shown in FIG. 5B, in the WFOV mode, the FLIR relay / FOV assembly 407 includes not only the single lens 412a and the third lens set 504, but also the first lens set 501 and the second lens set 502 as described above. Is used. Again, the third lens set 504 is shown as a single lens 412b in FIG. Unlike the first lens set 501 and the second lens set 502, the single lens 412a and the third lens set 504 are not fixed with respect to the FLIR optical path. When the FLIR relay / FOV assembly 407 transitions to the WFOV mode, they rotate and enter the FLIR optical path, and when the FLIR relay / FOV assembly 407 returns to the NFOV mode, they are out of the FLIR optical path. The single lens 412a and the third lens set 504 are rotated to enter and exit the FLIR optical path by a WFOV switch (not shown). The single lens 412a and the third lens set 504 together with the fixed lens described above produce a minimum afocal magnification and become a large FOV.
Both the fixed lens and the rotating lens are arranged around the intermediate focal plane along the IR optical path so that a real entrance pupil and a real exit pupil are generated in both the NFOV mode and the WFOV mode. Thereby, regardless of whether the system is operating in NFOV mode or WFOV mode, to the same common aperture 401 in the pitch / yogimbal assembly 460 without causing a function smaller than 2.0 micron wavelength. IR energy can be passed.
FIG. 4 also shows the optical path for laser energy. During propagation of the laser energy, the collimated laser energy exits the transmitter 450 and enters the compensator assembly 452. The compensator assembly includes a compensator unit 454 and a pair of risley prisms R1. The compensator unit 454 adjusts the phase relationship between the orthogonal polarization states of the incident laser beam, and as described above, its purpose is to ensure that a sufficient amount of laser energy passes through the beam splitting cube 470 as the deroll prism 414 rotates. It is to guarantee. The collimated laser energy is then passed from the compensator unit 454 to the Risley prism R1, which is used to steer the laser LOS. The compensator assembly 452 will be described later.
The laser energy exits the compensator assembly 452 and is reflected toward the FSM 415 by the dichroic D3. As noted above, the FSM 415 further minimizes both laser LOS and FLIR LOS jitter to further ensure that the laser energy is accurately directed to the desired target.
The laser energy is then passed to a deroll prism 414 where it is internally reflected as described above for IR energy. Upon exiting deroll prism 414, the laser energy is reflected by dichroic D 2 through integrated optics assembly (COA) 456 and laser relay / focus assembly 458. Finally, the laser energy strikes the dichroic D1 and is reflected to the pitch / yogimbal assembly 460 and then directed along the common pitch axis through the common aperture 401 toward the desired target.
Laser energy returning from the desired target enters the sensor through a segmented window (not shown) and then passes through the common aperture 401. It then reaches D1 along the same optical path as the IR energy (ie, along the common pitch axis) where it is reflected through the laser relay / focus assembly 458 and enters the COA 456. Next, depending on the state of the LST / LRR switch (LRS) 466, the COA 456 directs the return laser energy back to the laser range receiver (LRR) 462 or the laser spot tracker (LST) 464.
The laser subsystem components, including laser transmitter 450, compensator assembly 452, COA 456, LRR 462, LST 464, and laser relay / focus assembly 458 will now be described in detail.
In the preferred embodiment, laser transmitter 450 is a diode pumped solid state ND: YAG laser. Diode pump transmitters are preferred over flash lamp pump lasers of the same light output because of their high reliability, low power requirements and long lifetime. Laser transmitter 450 uses an optical parametric transmitter to generate a 1.57 micron training wavelength in addition to a 1.064 micron tactical wavelength. It has a beam quality (ie brightness) better than 9 mm-mrad when the output of the laser transmitter 450 is measured at the laser output port, and more than 90 mJ at 1.064 microns and from 26 mJ at 1.57 microns. Parts can be obtained so that more can be transmitted. In addition, the laser transmitter output can be driven by a pulse period modulation (PIM) code, i.e., with a pulse repetition frequency (PRF) up to 20 Hz with a pulse width of 20 nanoseconds.
The LRR462 consists of a low noise InGaAs receiver that responds to both 1.064 micron tactical wavelength and 1.57 micron training wavelength. The LRR 462 is remotely located from the laser and uses the low level backscatter of the emitted laser pulse as a trigger for its timing circuit. The LRR 462 is remotely located from the laser transmitter 450 and is polarized and separated by the COA 456 to prevent the emitted laser energy from being specularly reflected into the LRR 462. The energy level of the outgoing laser pulse is several orders of magnitude higher than the energy level of the return laser pulse. Even if the energy of the emitted laser pulse is very small, if it is reflected and enters the LRR 462, the LRR is saturated and the return pulse cannot be detected in a timely manner. The LRR 462 can measure a distance from about 0.463 km (0.25 nautical miles) to about 37 km (20 nautical miles) with an accuracy of ± 3 m.
In the preferred embodiment of the present invention, the LST 464 uses a quad cell silicon detector to obtain a laser spot track.
Laser relay / focus assembly 458 provides two functions. First, it relays the transmission and reception of laser energy between the dichroic D1 and the COA 456. Second, the laser energy focus is maintained over temperature and altitude by translating a set of focus lenses FL3. The laser relay / focus assembly 458 translates the focus lens set FL3 along a linear bearing by using a motor driven lead screw and monitoring position feedback with a potentiometer. In effect, the laser energy is focused inside a vacuum cell (not shown) incorporated within the laser relay / focus assembly housing. Laser energy enters and exits the vacuum cell through a set of vacuum cell windows 468.
As described above, the compensator assembly 452 includes a pair of risley prisms R1 and a compensator unit 454, and performs three main functions. First, the compensator assembly 452 can shift the laser LOS to adjust the laser to FLIR boresight. Second, the compensator assembly 452 can adjust the polarization of the laser energy to compensate for the perturbations introduced by the deroll prism 414. Third, as described below, the compensator assembly can adjust the polarization of the laser energy to alternate the intensity of the laser energy during the internal boresighting procedure. These tasks are performed by the compensator unit 454.
As shown in FIG. 6, the compensator unit 454 includes two optical wave plates, a λ / 4 wave plate 605 and a λ / 2 wave plate 610, which monitor the deroll angle (ie, the angle of the deroll prism 414). Under the control of a microprocessor-based circuit (not shown), it is rotated individually as a function of the deroll angle. Both λ / 4 wave plate 605 and λ / 2 wave plate 610 are individually mounted in the same ball bearing support gear drive housing, each with its own drive mechanism including servo motor and position feedback resolver. .
FIG. 6 shows that the outgoing laser beam 615 is linearly polarized when leaving the laser transmitter 450. However, to ensure that the laser beam is linearly polarized upon exiting the deroll prism 414, the λ / 4 wave plate 605 can be rotated to change the ellipticity of the laser beam polarization, and the λ / 2 wave plate. 610 can be rotated to rotate the orientation of the ellipse. By ensuring that the laser beam is linearly polarized as it exits deroll prism 414, the laser beam is linearly polarized as it passes through polarizing beam splitter cube 470, and the energy level of the propagating laser beam is sufficient to illuminate the desired target. Guaranteed to be sufficient. It should be understood that the order in which the laser beam meets the λ / 4 wave plate 605 and the λ / 2 wave plate 610 is important.
The COA 456 is an electromechanical / optomechanical assembly that directs the returning laser energy beam back into the LRR 462 or LST 464. The COA 456 also helps maintain proper laser pair LRR boresight and proper laser pair LST boresight.
The laser energy is randomly polarized as it returns from the desired target to the system. In the COA 456, the randomly polarized laser energy is again directed into the λ / 2 wavelength plate 466 by the polarizing beam splitter 470. In the preferred embodiment, the λ / 2 wave plate 466 rotates the polarization of the laser energy such that the laser energy is adjusted, i.e., split between, relative to the LRR detector, LST detector. Therefore, the λ / 2 wave plate 466 functions as an LRR / LST switch.
After the λ / 2 wavelength plate 466 polarizes the laser energy, the laser LOS in the COA 456 is shifted using the same pair of Risley prisms R2 as the pair of Risley prisms R1, and the laser pair LRR or the laser pair LST The boresight is adjusted. Both Risley prism R1 and λ / 2 wave plate 466 have their own drive mechanisms including servo motors and position feedback resolvers.
After passing through the pair of Risley prisms R 2, the laser energy collides with the polarization beam splitter 472. Polarization beam splitter 472 directs laser energy to LRR 462, LST 464, or both again, depending on the rotational state of λ / 2 wave plate 466 (ie, LRR / LST switch).
In order to stabilize, modify and control the laser LOS and FLIR LOS, the present invention utilizes a LOS / servo subsystem. The LOS / servo subsystem helps to minimize both FLIR LOS and laser LOS dynamic jitter. FIG. 7 illustrates a LOS / servo sub that includes a 6-axis gimbal assembly 705, several single-phase, two-phase and three-phase servo motors 710, a power amplifier 715, various rate and position sensors 720, a digital interface 725, and a digital processor 730. 1 is a diagram of a system.
The six-axis gimbal assembly 705 includes a pitch / yogimbal assembly 460, a roll gimbal (not shown), a deroll assembly 413, and an FSM 415. Pitch / yogimbal assembly 460, deroll assembly 413, and FSM 415 are all mounted on a passive insulating light bed 490 supported by a roll gimbal. The insulated light bed 490 attenuates high frequency vibrations and minimizes structural bending (ie, low to intermediate frequency vibrations) that can contribute significantly to LOS and dynamic alignment errors.
Pitch / yogimbal assembly 460 includes an internal gimbal, which includes a set of lenses 410a, 410b, 410c and a gimbal mirror 403, as shown in FIGS. 5A and 8. It should be understood that the lens set including the lenses 410a, 410b, and 410c is shown as a single lens 401 in FIG. The lenses 410a, 410b, and 410c are optical elements that constitute a common pitch / yaw focal (that is, the above-described common aperture). The internal gimbal provides limited rotational motion (approximately 5 °) about the yaw axis. The pitch / yogimbal assembly 460 also has an external gimbal that includes four lenses 815 (ie, a pitch gimbal). The external gimbal provides a full 360 ° rotation about the pitch axis 405. Furthermore, there is only one pitch axis interface between the pitch / iodine gimbal assembly 460 and the host platform, and thus there is only a common pitch bearing set for laser optics and FLIR optics. The pitch / yogimbal assembly 460 utilizes two separate servo motors (not shown).
Roll gimbal and deroll assemblies 413 provide LOS pointing and image roll stabilization, respectively. In particular, deroll assembly 413 provides excellent roll stabilization and horizontal stabilization for FLIR images when the system rotates about the roll axis and / or the gimbal assembly rotates about the pitch axis. In practice, image derolling is accomplished by a derolling prism 414 mounted in a cylindrical housing that rotates on a low friction, preload, combination bearing. The rotary assembly is driven directly by a pancake torque motor, and the position is sensed via a pancake resolver mounted coaxially with the motor.
FSM assembly 416 includes piston probe electronics 930 and electromagnetic assembly 900. As shown in FIG. 9 (see also US Pat. No. 5,550,669), FSM electromagnetic assembly 900 is an electromagnetic device that rotates FSM 415 about pitch axis 910 and yaw axis 915. In the preferred embodiment, FSM 900 provides excellent stabilization against FLIR LOS and laser LOS. It is accomplished by measuring the pitch and yam gimbal motion with the IMU 720 and applying a position correction command to the FSM position servo motor 917. The FSM servo motor position command is then controlled by the FSM servo electronics assembly 920. FSM position feedback is provided by a set of position probe 925 and position probe electronics assembly 930.
LOS control sensor 720 provides digital information to LOS control software that resides within LOS control processor 730. The LOS control processor generates LOS control signals that control both the FLIR LOS and the laser LOS. The LOS control sensor 720 includes an angular roll rate sensor that provides digital roll rate information, a resolver that provides gimbal angular position, a position probe that provides position information for the FSM 415, and an inertial measurement unit (IMU) assembly (not shown). It is out. The IMU assembly includes a three axis fiber optic gyro (FOG) and a three axis acceleration watch. The IMU provides incremental angular information as a function of inertial angular velocity and incremental velocity information as a function of inertial linear acceleration.
While the LOS servo subsystem helps to minimize dynamic alignment errors, the present invention includes an internal boresight module (see FIG. 4) for minimizing fixed alignment errors between various internal lines of sight. BM) 474. The BM 474 is optically connected to the pitch / yogimbal assembly 460 and is used with the boresight control algorithm to align the internal line of sight with each other. Internal line of sight includes FLIR NFOV LOS, FLIR WFOV LOS, laser transmitter LOS, LRR LOS, and LST LOS.
FIG. 10 shows the details of BM474. The BM 474 contained within the housing 476 includes an aperture window 1005 through which IR and laser energy passes. The BM 474 also includes a Cassegrain optical system including a primary mirror and a set of secondary mirrors 1006 and 1007 that direct IR and laser energy into and out of the BM 474, respectively. In addition, the BM 474 includes a reticle 1010, a laser filter 1025 that prevents laser energy from passing through an opening in the reticle 1010, an IR / visible light source 1015, a laser detector 1020, and a laser source 1030.
During the FLIR LOS alignment procedure described below, IR / visible light source 1015 emits both IR energy and visible light energy, which passes through diffuser 1016, dichroic 1017, and optical lens 1018. The energy is then reflected from the second dichroic 1019 via the reticle 1010 and is directed out of the BM 474 by the primary and secondary mirrors 1006 and 1007.
As described below, during the laser LOS alignment procedure, laser energy from the laser transmitter 450 is directed into the BM 474 by the primary and secondary mirrors 1006 and 1007. The laser energy then passes through reticle 1010 and enters laser detector 1020 via dichroic 1019 and optical lens 1024.
During the LRR and LST boresight procedures, the laser source 1030 emits laser energy that passes through the optical lens 1031. Laser energy is reflected from dichroic 1017, passes through optical lens 1018, is reflected from second dichroic 1019, passes through reticle 1010, and exits BM 474. In the preferred embodiment, laser source 1030 is a laser emitting diode (LED).
Reticle 1010 includes a reticle pattern 1100 as shown in FIG. Reticle pattern 1100 has a number of unusually shaped apertures through which laser or IR energy passes during various LOS alignment processes. As shown in FIG. 11, the various openings are symmetrically arranged around a 1064 microradian circular region. At the very center of this circular area is a 60 microradian hole that coincides with the center of the reticle pattern 1100.
The first set of openings in reticle pattern 1100 includes five octagonal openings 1105a-e. These openings allow IR energy to pass through the reticle 1010 during the FLIR LOS alignment procedure. As described below, the first four apertures 1105a-d are used to align the FLIR LOS symmetrically with respect to the center of the FLIR focal plane array. The fifth aperture is used to rotationally align the FLIR LOS with respect to the FLIR focal plane array.
The reticle pattern 1100 also has a number of checkered openings 1110. The checkered aperture is used with special signal processing software. The software then controls the procedure for imaging the FLIR image prior to the FLIR alignment procedure. The large checkered aperture is specifically used to image a WFOV FLIR image, and the small checkered aperture is specifically used to image an NFOV FLIR image. The signal processing software that uses the checkered pattern opening will be described later.
Finally, the reticle pattern 1100 includes four wedge-shaped openings 1115. Unlike the octagonal aperture described above, the wedge-shaped aperture 1115 can pass laser energy from the laser transmitter 450 to the reticle 1010 during the laser LOS alignment procedure.
Next, a method for boresighting the FLIR LOS, the laser transmitter LOS, and then the LRR LOS and LST LOS will be described in detail with reference to FIG. Initially, a boresight command is generated according to block 1205. Next, as shown in block 1210, the pitch gimbal rotates 165 °. By rotating the pitch gimbal by 165 °, the BM474 becomes an integral part of the FLIR optical path and the laser optical path. Next, a signal processing routine is activated in accordance with block 1215 that controls the servo motor to move various IR and laser optics during the boresight procedure. The NFOV FLIR and WFOV FLIR images are then imaged on the focal plane array 480 in the FLIR detector / cooler assembly 427 according to blocks 1220 and 1225, respectively. According to block 1230, the FLIR LOS alignment procedure is initiated once the NFOV and WFOV FLIR images are imaged. As described above, the FLIR LOS is aligned with the center of the focal plane array 480 using the IR / visible light source 1015, the reticle pattern 1100, and in particular the octagonal apertures 1105a-e in the reticle pattern 1100. The FLIR LOS alignment procedure is performed by first irradiating with an IR / visible light source 1015. IR and visible light energy pass through five octagonal apertures 1105a-e in reticle pattern 1100. The energy passing through each of the five octagonal apertures 1105a-e illuminates five corresponding regions on the focal plane array 480. When the roll servo is fixed, the first four irradiation areas on the focal plane array 480 corresponding to the first four octagonal apertures 1105a-e correspond to the pitch and low servo by the signal processing routine described above. You are instructed to rotate the pitch and yam gimbal until they are placed symmetrically around the center. Next, if the pitch and yaw servos hold their positions, the signal processing routine and the deroll servo use the fifth illumination area on the focal plane array 480 corresponding to the fifth octagonal aperture 1105e, Roll alignment of reticle pattern 1100 with respect to focal plane array 480 is performed to complete FLIR LOS alignment with BSM.
Once the FLIR LOS is aligned, the laser LOS is aligned as shown at block 1235. In order to align the laser LOS, the laser transmitter 450 begins to generate a continuous pulse stream at a nominal rate of 20 Hz. As described above, the laser filter 1025 passes laser energy from the laser transmitter 450 only through certain openings in the reticle pattern 1100, in particular, the 60 ° wedge-shaped openings 1115 and 60 microradians holes at the center of the reticle pattern 1100. As shown at position 1305 in FIG. 13, when the laser transmitter 450 begins transmitting, the laser spot may not coincide with one of the four wedge-shaped openings 1115. To align the laser spot with one of the four wedge-shaped apertures 1115, the boresight algorithm instructs the pair of Risley prisms R1 to move the laser spot to various positions within the search pattern 1310. Finally, the laser spot coincides with one of the wedge-shaped openings 1115, as indicated by position 1315. As a result, it is detected by the laser detector 1020. Next, as shown at position 1320, the boresight algorithm instructs Risley pair R1 to move the laser spot to the nearest radial edge of the corresponding wedge-shaped aperture. Based on the angle of the radial edge, the boresight algorithm can determine the direction in which the laser spot must be moved to reach the 60 microradian hole at the center of the reticle pattern 1100. In FIG. 13, the moving direction of the laser spot is indicated by continuous laser spot positions 1320, 1325 and 1330.
When the laser spot reaches the inner edge of the corresponding wedge-shaped opening, it is approximately 500 microradians from the center of the reticle pattern 1100. Next, as shown in FIGS. 14A to 14F, the boresight algorithm instructs Risley pair R1 to move the laser spot to several grating positions according to one of several possible rectangular grating patterns, Each rectangular grid pattern includes a portion of a 1064 microradian circular region centered on reticle pattern 1100 that includes exactly the center 60 microradian hole. The particular grating pattern used depends on the angle of the radial edge used as a guide to move the laser spot toward the center of the reticle pattern 1100. In the previous example, the laser spot moved along the lower edge of the wedge-shaped opening 1350 toward the center of the reticle, and was positioned at the position 1405 shown in FIG. 14A when reaching the inner edge of the shaped opening 1350. Accordingly, the rectangular lattice pattern shown in FIG. 14A is used as a guide for finely adjusting the alignment of the laser spot with respect to the center of the reticle pattern 1100.
Prior to Risley pair R1 moving the laser spot to each grating pattern position associated with the grating pattern shown in FIG. 14A for example, the boresight algorithm directs polarization compensator 454 to attenuate the amount of laser energy delivered to the BSM. The The laser energy is attenuated to prevent the focused laser energy from damaging the reticle and to adjust the energy level to be within a range detectable by the laser detector 1020. After attenuating the laser energy, Risley pair R1 moves the laser spot to each grating pattern position. The position corresponding to the maximum laser energy detection detected by the laser detector 1020 is identified as the best laser LOS boresight alignment position.
Next, according to block 1237, the image illuminated on the focal plane array 480 is rotated 180 °. This is done by rotating the deroll prism by 90 °. The FLIR LOS procedure according to block 1230 and the laser LOS procedure according to block 1235 are then repeated. By repeating these procedures, any alignment error between the FLIR LOS and the laser LOS due to the fact that the IR energy and laser energy travel in different optical paths is calibrated by the boresight algorithm.
The IR energy passes through the octagonal aperture 1105 and they are symmetrically arranged around the 60 microradian hole in the center of the reticle pattern 1100, so alignment of the laser LOS with respect to the 60 microradian hole causes the FLIR LOS and laser LOS to be aligned. Precise alignment is achieved. The laser LOS vs. FLIR LOS boresight algorithm described above is accurate to within 42 microradians and the laser beam diffusion is 120 microradians or less.
Finally, LRR 462 and LST 464 are aligned according to blocks 1240 and 1245, respectively. To align LRR 462 and LST 464, laser transmitter 450 is turned off and laser diode 1030 is turned on. Laser energy emitted from laser diode 1030 is sent through a 60 microradian hole located at the center of reticle pattern 1100. Next, energy from the laser diode 1030 is passed into the COA 456 along the laser beam path. The COA 456 then directs laser energy into the LRR 462. The LRR LOS is aligned by rotating the Risley prism 466 pair. The COA 456 then directs the laser energy into the LST 464 and the LST LOS is similarly aligned.
In an alternative embodiment, the electro-optic subsystem employs a combined LST / LRR assembly 1505, as shown in FIG. When LST and LRR are parts of a single unit, there is no longer any need to direct laser return energy into LRR and LST. Thus, in an alternative embodiment, the COA 456 is greatly simplified. In particular, the LRR / LST switch 466, BSC 472, and two mirrors are removed, thereby reducing the complexity of the COA 456. Furthermore, by combining LRR and LST, there is no longer a need to aim LRR and LST separately. There is also additional space available for the television camera 1510 by combining the LRR and LST and by removing the LRR / LST switch 466, BSC 472, and the two aforementioned mirrors, and this camera Can be used to represent a near IR (infrared) image of AOI.
FIG. 16 shows the configuration of the combined LRR / LST assembly 1505. The LST is preferably a quad cell photosensitive device in which each cell outputs an electrical signal 1605a through 1605d, each of which is proportional to the amount of laser energy that illuminates the corresponding cell. Quad cells LST are generally well known to those skilled in the art. The LRR 1610 is a pin diode positioned directly in the center of the quad cell device at the intersection of two high impedance buffer regions 1615a and 1615b, which electrically insulate the quad cells from each other. The LST quad cell is housed in a hybrid together with an integrated LRR pin diode. A preamplifier and a postamplifier for both the LST sensor and the LRR sensor are co-located in the hybrid.
Segmented window and EMI grid
In the preferred embodiment, the opto-electrical subsystem described above is housed in a pod assembly, and the front (ie, shroud) of the assembly is shown in FIG. The pod assembly is connected to the host platform. For example, the pod assembly is suspended from a pylon on the wing of a combat aircraft such as F-15, F-16, or F-18. The shroud assembly physically protects the opto-electronic equipment described above, while laser energy and IR (infrared) energy pass through the segmented window 1705 into and out of the Shland assembly.
The window has four segments or panels 1705a to 1705d. These segments are designed to optimize the air-optical performance of the system. For example, these segments or panels reduce or prevent energy from reflecting back into the opto-electronic sensor described above. These panels also optimize the modular transfer function (MTF) versus gimbal angle at mission critical pitch angles of 0 ° and 20 at 20,000 AGL (ground altitude). In addition, these window panels support sensor scanning over + 35 ° to -155 ° in pitch, 360 ° in rolls, and ± 5 ° in yaw. In addition, segmented windows are employed instead of shroud followers. The shroud follower requires more hardware and exhibits a large radar cross section (RCS) and aerodynamic drag coefficient.
The four window panels are composed of the substrate material and the three required coatings. These coatings are electromagnetic interference coatings, durable anti-reflection (DAR) coatings, and internal anti-reflection (IAR) coatings. A number of different materials can be used for the substrate material. For example, “Cleartran” (multispectral ZnS) is commonly used for this purpose. This exhibits good multispectral characteristics, but is not very durable. In contrast, sapphire is very durable and it exhibits good transmission properties in the mid-wave range (ie, ˜3 to 4.5 micrometers). Thus, in the preferred embodiment, sapphire is used as the substrate material.
Each of the aforementioned panels also includes an EMI (electromagnetic interference) grating. In general, EMI gratings reduce the exposure of sensors and sensor electronics to large EM (electromagnetic) fields that may be present in the operating environment. For example, EMI gratings minimize the impact of EM field energy on system electronics (eg, IMU (Inertial Measurement Unit), FLIR (Forward Monitoring Infrared Radar) Analog Electronics, FSM (Flying Point Microscope) probe) These system electronics fire from a source (ie, radar) installed on a host platform (eg, F-16, F-18) that may be used to limit or eliminate Sensitive to this type of radiation. This grid may also minimize EM energy from the electronics installed in the pod assembly, which can adversely affect the electronics installed on the host platform.
In general, EMI gratings are well known to those skilled in the art. However, the leading EMI grid employs a square pattern or a polar (concentric) pattern. These preceding EMI grid patterns generate unacceptable artifacts on the FLIR image. For example, a square grid pattern concentrates energy along four lines that radiate from an off-axis light source, resulting in an energy artifact 1800 as shown in FIG. This energy appears on the FLIR image and is unacceptable.
In the preferred embodiment, an EMI grid pattern including a circular array is employed as shown in FIG. This EMI grating pattern guides or diffracts stray light radially rather than along a particular axis, thus reducing or eliminating unwanted light energy artifacts on the FLIR image. In the arrangement of FIG. 19, each circle preferably has a line width of 5 micrometers and a diameter of 320 micrometers. The repeat offset (ie, the distance between the centers of two adjacent circles) is 315 micrometers, where there is a 100 percent overlap of the circular line at the tangent point of the circle. The particular EMI grating described above represents just one exemplary embodiment of an EMI grating. As those skilled in the art are aware, other dimensions may be employed without exceeding the intended range of the EMI grid pattern described above.
The EMI grid of the present invention is attached to the segmented window panel in exactly the same manner as in the prior art. First, three metal layers are applied to the window: an inner adhesive layer, an intermediate conductive layer, and an outer protective layer. Different metals may be used for each layer, however, typically chromium is used for the inner layer, gold for the intermediate layer, and titanium for the outer layer. A “mask” is placed over the window material and then exposed to ultraviolet light. Ultraviolet light removes the metal layer from areas not protected by the mask. When the mask is removed, the EMI grid is secured to the window.
Fault isolation
The electro-optic subsystem includes a number of servo systems. Each servo system includes one or more amplifiers and a single phase, two phase, or three phase servo motor. These servo systems provide the power or mechanical force necessary to rotate and / or translate the various lenses, prisms, mirrors, and wave plates in the electro-optic subsystem. The servo system includes a two-phase brushless DC motor and a pulse width modulation (PWM) amplifier servo system (ie, pitch servo system and deroll servo system). Also included are three-phase brushless DC motors, PWM amplifier servo systems (ie, roll servo systems). In addition, several single-phase motors, linear amplifier servo systems (ie, various laser and FLIR focusing assembly servo systems, thermal reference servo systems, FSM servo systems, Risley prism servo systems, yaw servo servo systems) Systems, and control servo systems, inlet and outlet airflow servo systems) are also included.
In order to detect the presence of a fault condition in any of the servo systems identified above, the present invention allows a fault condition to occur in the amplifier portion of a given servo system or in the motor portion of a given servo system. Provides fault isolation capability that can be determined whether or not. By isolating faults to the amplifier or motor portion of a given servo system, the need to remove or replace the entire servo system is avoided. Instead, only the amplifier part or motor part containing the fault condition need be removed and / or replaced. The prior art has required test loads and switching means that cause these loads to enter and exit the circuit. This technique does not require a test load.
FIG. 20 is a circuit diagram of a two-phase brushless DC motor, PWM amplifier servo system circuit 2000. Under normal conditions, the circuit 2000 controls the commutation of the current through the windings of the two-phase motor M, which in turn controls the rotation of the motor shaft. In particular, the two-phase brushless DC motor, PWM amplifier servo system circuit 2000 includes two PWM amplifiers A and B. Each of these amplifiers A and B has two upper excitation transistors T1 and T2 and two lower excitation transistors T3 and T4. In order to rectify the current through the motor windings and thus rotate the shaft of the two-phase motor M, the upper excitation transistor in each amplifier is placed under the other side of the same amplifier as will be understood by those skilled in the art. It is necessary to pair with a side excitation transistor. For example, in amplifier A, transistor T1 will be paired with transistor T4 and transistor T2 will be paired with transistor T3. Similarly, in amplifier B, transistor T1 will be paired with transistor T4 and transistor T2 will be paired with transistor T3. Thus, by activating each pair of excitation transistors alternately in amplifier A and then in amplifier B (a technique known as complementary switching), a two-phase motor, PWM amplifier servo system circuit 2000 can be A constant torque can be maintained on the M axis.
The two-phase motor, PWM amplifier servo system circuit 2000 also includes a control circuit 2005. The control circuit 2005 specifically controls the timing of the complementary switching process (ie, when each excitation transistor pair is activated) to control the load current using a standard feedback configuration. In the present invention, the control circuit 2005 is realized by a field programmable gate array (FPGA). However, as those skilled in the art will readily appreciate, the control circuit 2005 can be implemented using other forms of logic without departing from the spirit of the present invention.
In addition, each amplifier A and B in the two-phase motor, PWM amplifier servo system circuit 2000 includes two upper current sensing resistors R1 and R2. Upper current sensing resistors R1 and R2 monitor the amount of current flowing through upper excitation transistors T1 and T2, respectively. In addition, each amplifier A and B includes a lower current sensing resistor Ro. The lower current sensing resistor Ro is used to monitor the current flowing through the lower excitation transistors T3 and T4. It is important to note that in the present invention amplifiers A and B include insulated gate bipolar transistors (IGBTs). IGBTs are used in part because their use simplifies the design of the control circuit 2005 given the 270 volt power supply Vcc, as will be readily understood by those skilled in the art. However, as those skilled in the art will also appreciate, solid state switches such as MOSFETs and bipolar transistors can also be employed.
In general, the two-phase, PWM amplifier servo system fault isolation process works as follows. Complementary switching of the transistor pair is disabled. A control circuit 2005 activates a pair of excitation transistors, eg, T1 and T4 in amplifier A. The control circuit 2005 then deactivates this excitation transistor pair and replaces each of the other excitation transistor pairs: T2 and T3 in amplifier A, T1 and T4 in amplifier B, and amplifier B. Activates and deactivates T2 and T3. Under normal operating conditions (ie, when no fault condition exists), activation of each excitation transistor pair causes a specific amount of current to flow through the corresponding upper current sensing resistor R1 or R2 to the corresponding upper excitation transistor T1 or It flows through T2, through the corresponding servo motor winding, through the corresponding lower excitation transistor T3 or T4, and finally through the lower current sensing resistor Ro. In fact, if no fault condition exists, a known voltage drop will always appear across the upper current sensing resistor R1 or R2 and across the lower current sensing resistor Ro. However, if a fault condition exists, the amount of current flowing through the upper current sensing resistor R1 or R2 and / or across the lower current sensing resistor Ro will be significantly affected. By monitoring the amount of current flowing through a resistor, such as each of the excitation transistor pairs identified above, as described in more detail below, the following single point fault conditions: motor winding short circuit, motor The winding open circuit, amplifier short circuit, motor winding ground fault, and amplifier open circuit can be separated.
FIG. 21 is a flow diagram illustrating the steps of a two phase motor, PWM amplifier fault isolation process, according to a preferred embodiment of the present invention. According to FIG. 21, the control circuit 2005 begins the process by disabling complementary switching as indicated by block 2101. Once complementary switching is disabled, the control circuit 2005 is free to activate and deactivate the excitation transistor pair for fault isolation purposes.
After complementary switching is disabled, control circuit 2005 generates a positive control current command + I, as indicated by block 2105. The positive control current command activates the transistor pair including the excitation transistors T1 and T4. As indicated by decision block 2110, the control circuit 2005 determines whether the magnitude of the current flowing through the upper current sensing resistor R1 is normal or whether it exceeds a normal amount, ie, it is a predetermined threshold. Determine whether the value is exceeded. If the magnitude of the current flowing through the upper current sensing resistor R1 is excessive, according to the “YES” path exiting the decision block 2110, the control circuit 2005, as indicated by the decision block 2112, will detect the lower current sensing. It is determined whether the magnitude of the current flowing through resistor R0 is equal to zero. If the current flowing through the lower current sensing resistor R0 is equal to zero, the control circuit 2005 will generate an amplifier SHUTDOWN command as indicated by block 2113. In order for the control circuit 2005 to determine that the amount of current is equal to zero, the fault condition is identified as a motor winding ground fault, as indicated by block 2114. If the control circuit 2005 determines that the current flowing through the lower current sensing resistor is not equal to zero, according to the “NO” path exiting the decision block 2112, the control logic 2005, as indicated by block 2115, Nevertheless, it will generate an amplifier SHUTDOWN instruction. However, it is still not determined whether the fault condition caused by the excessive amount of current through the upper current sensing resistor R1 is due to a short circuit in the motor winding or a short circuit in the amplifier. Thus, the control circuit then generates a negative current command -I, as indicated by block 2120, which activates the transistor pair including the excitation transistors T2 and T3. The control circuit 2005 now determines whether the amount of current flowing through the upper current sensing resistor R2 is excessive, ie, if it exceeds a predetermined threshold, as indicated by block 2125. . If the amount of current flowing through the upper current sensing resistor R2 exceeds the threshold value, then following the “YES” path exiting decision block 2125, the control circuit 2005 once again, as indicated by block 2130, the amplifier SHUTDOWN. Generate an instruction. Since an excessive amount of current has been drawn through both upper current resistors RI and R2, the fault condition is identified as a servo motor winding short as indicated by step 2135. However, if the magnitude of the current flowing through the upper current sensing resistor R2 does not exceed a predetermined threshold value, then follow the “NO” path from decision block 2125, ie, the upper current sensing resistor will draw an excessive amount of current. Only if, as indicated by step 2140, the fault condition is identified as an amplifier short.
If the positive current command + I is generated according to block 2105, the magnitude of the current flowing through the upper current sensing resistor R1 exceeds the predetermined threshold value according to the “NO” path from which the control circuit 2005 exits the determination block 2110 If not, the control circuit 2005 determines whether the magnitude of the current flowing through the lower current sensing resistor Ro is equal to zero, as shown in decision block 2145. If the magnitude of the current flowing through the lower current sensing resistor is equal to zero, the control circuit 2005 generates a negative command current -I as indicated by block 2150 and indicated by decision block 2155. As described above, it is determined whether or not the magnitude of the current flowing through the upper current sensing resistor R2 exceeds a predetermined threshold value. If the magnitude of the current flowing through the upper current sensing resistor R2 exceeds a predetermined threshold, following the “YES” path exiting decision block 2155, the control circuit 2005, as indicated by block 2160, will amplify the amplifier SHUTDOWN. An instruction will be generated. Since an excess of current was detected through only one of the upper current sensing resistors, ie, R2, the fault condition is identified as an amplifier short, as indicated by step 2165.
However, if the magnitude of the current flowing through the upper current sensing resistor R2 does not exceed the predetermined threshold, the control circuit 2005 follows the “NO” path from decision block 2155 as indicated by decision block 2170. Then, it is determined whether the magnitude of the current flowing through the lower current sensing resistor is equal to zero. If the current flowing through the lower current sensing resistor Ro is equal to zero, ie both T1 and T4 are activated and T2 and T3 are activated, both R1 and R2 draw a normal amount of current and Ro If does not draw current, the fault condition is identified as an open motor winding, as indicated by step 2175. If the lower current sensing resistor Ro is not equal to zero, then both resistors R1 and R2 follow the “NO” path exiting decision block 2170, ie, only when T1 and T2 are activated. The amount is being drawn and Ro is not drawing current, and, as indicated by step 2180, the fault condition is identified as an amplifier open.
However, if the magnitude of the current flowing through the lower current sensing resistor Ro is not equal to zero, according to the “NO” path exiting decision block 2145, the control circuit 2005 is negative as indicated by block 2185. A command current -I is generated which activates the transistor pair including the excitation transistors T2 and T3. The control circuit 2005 then determines whether the amount of current flowing through the upper current sensing resistor R2 exceeds a predetermined threshold, as indicated by decision block 2190. If the magnitude of the current flowing through the upper current sensing resistor R2 exceeds the threshold value, then follow the “YES” path from decision block 2190, ie, an excess of current flows through only one of the upper current sensing resistors, The control circuit 2005 generates an amplifier SHUTDOWN instruction as indicated by block 2160 and the fault condition is identified as an amplifier short as indicated by step 2165.
However, if the magnitude of the current flowing through the upper current sensing resistor R2 does not exceed the predetermined threshold, the control circuit 2005 follows the “NO” path exiting from decision block 2190 as indicated by decision block 2195. And determine whether the magnitude of the current flowing through the lower current sensing resistor Ro is equal to zero. If the current flowing through the lower current sensing resistor Ro is equal to zero, following the “YES” path exiting decision block 2195, ie, R1 and R2 are normal amounts of current only when T2 and T3 are activated. And the amount of current through Ro is zero, and as indicated by step 2180, the fault condition is identified as an amplifier open circuit. If the magnitude of the current flowing through the lower current sensing resistor Ro is not equal to zero, follow the “NO” path from decision block 2195, ie, when either excitation transistor pair is activated, R1 and R2 Is drawing a normal amount of current and Ro is not zero, the control circuit 2005 will indicate that no single point fault is identified as indicated by step 2199.
Since the fault isolation process described above is only suitable for one phase of a two-phase motor, PWM amplifier servo system, the above fault isolation process is applied to the second phase (ie, the second phase of the motor and the second amplifier). Can be repeated) to isolate faults between In fact, if a three or more phase servo system is employed, the above process can be repeated as many times as there are phases present. Table I summarizes the two-phase, PWM amplifier fault isolation process according to a preferred embodiment of the present invention.

FIG. 22 is a circuit diagram of a three-phase motor, PWM amplifier servo system circuit 2200. A three-phase motor, PWM amplifier servo system circuit 2200 includes a phase exciter A including an upper excitation transistor T1 and a lower excitation transistor T4, a phase exciter B including an upper excitation transistor T2 and a lower excitation transistor T5, and an upper 25 A PWM amplifier having a phase exciter C including an excitation transistor T3 and a lower excitation transistor T6 is employed. It is important to note that the excitation transistor is an IGBT. However, as will be appreciated by those skilled in the art, other solid state switches such as MOSFETs and bipolar transistors may also be employed. The three-phase motor, PWM amplifier servo system circuit 2200 also employs control logic 2205, which controls the following excitation transistor pairs: T1 and T5, T1 and T6, T2 and T6, T2 and T4, and T3. Controls the activation of T4 and T3 and T5. As those skilled in the art will readily appreciate, the activation of the transistor pair identified above is responsible for the three windings M of the three-phase motor M. _A , M _B , M _C To control the current through each of the. By appropriately activating each of the excitation transistor pairs identified above, the control logic 2205 can maintain a constant torque on the shaft of the three-phase motor M.
The control logic 2205 is controlled by software in the preferred embodiment of the present invention. Thus, the control logic 2205 receives a series of commutation codes referred to by those skilled in the art as a Hall code or a phase encoder feedback code. Each Hall code causes the control logic 2205 to activate the appropriate excitation transistor control lines 1,. For example, a Hall code responsible for activating a transistor pair including transistors T1 and T5 causes the control logic 2205 to activate the transistor control lines 1 and 5 regardless of the actual position of the motor shaft.
In accordance with the present invention, like the two-phase motor, PWM amplifier servo fault isolation process described above, the three-phase motor, PWM amplifier servo fault isolation process is an existing servo system circuit 2200 hardware for fault isolation purposes. Wear. Thus, the three-phase motor, PWM amplifier servo system circuit 2200 also includes three upper current sensing resistors R _A , R _B , R _C , And lower current sensing resistor R _out including. In general, fault isolation involves the supply voltage V to the servo system circuit. _CC Is applied, and then the upper current sensing resistor R _A , R _B , R _C And / or lower current sensing resistor R _out This is accomplished by isolating the single point fault condition by analyzing the amount of current flowing through it. Three-phase motor, PWM amplifier servo system fault isolation process, the following single point fault conditions: motor winding open circuit, motor winding short circuit fault, motor winding ground fault, amplifier excitation transistor open circuit, amplifier excitation transistor short circuit The amplifier output wire bond open circuit can be isolated.
FIGS. 23A and 23B are flowcharts illustrating the detailed steps of a three phase motor, PWM amplifier fault isolation process, in accordance with a preferred embodiment of the present invention. According to FIG. 23A, the control logic 2205 determines the desired excitation transistor pair by canceling the Hall code originally used to control the commutation of the current through the motor windings, and as indicated by block 2305. Start by generating a Hall code corresponding to (eg, upper excitation transistor T1 and lower excitation transistor T5). The Hall code will in turn cause the control logic 2205 to activate the appropriate transistor control line (eg, transistor control lines 1 and 5) corresponding to the desired transistor pair, as shown in block 2310.
Next, control logic 2205 determines the current flowing through the corresponding upper current sensing resistor (eg, upper current sensing resistor R, as indicated by block 2315). _A Or R _B Or R _C Current I through _A Or I _B Or I _C ) Must exceed a predetermined threshold T, where threshold T represents an otherwise excessive current. If the upper current sensing resistor is drawing an excessive amount of current, according to the “YES” path exiting decision block 2315, control logic 2205 generates a SHUTDOWN instruction, as indicated by block 2317, and An appropriate status flag is set as indicated by block 2320, thus indicating that activation of the current excitation transistor pair (eg, T1 and T5) has resulted in an overcurrent condition.
However, if the current excitation transistor pair activation did not result in SHUTDOWN, then according to the “NO” path exiting decision block 2315, control logic 2205, as indicated by decision block 2325, Activation is NO CURRENT condition (ie, output resistor R _out The current flowing through is zero). If activation of the excitation transistor pair results in a NO CURRENT condition, then according to the “YES” path exiting decision block 2325, control logic 2205 sets the appropriate status flag, as indicated by block 2330. Become.
The next step is to determine whether all six of the aforementioned excitation transistor pairs have been activated, as indicated by decision block 2335. If all six excitation transistor pairs are not activated, according to the “NO” path exiting decision block 2335, control logic 2205 generates a different Hall code to activate the next desired excitation transistor pair. It will be. According to the “YES” path from decision block 2335, once all six excitation transistor pairs have been activated, the status flag contains all of the information necessary to isolate the fault condition as described below.
After all six excitation transistor pairs have been activated, the status flag indicates that amplifier SHUTDOWN has occurred when two of the excitation transistor pairs have been activated, as indicated by decision block 2340, and these two effects. It may indicate that the excited transistor pair that was included included the same upper or lower driver transistor. For example, the first transistor pair may be a transistor pair including the excitation transistor pair T1 and T5, while the second transistor pair may be a transistor pair including the excitation transistor pair T1 and T6. Here, however, both excitation transistor pairs include the upper excitation transistor T1 from phase exciter A. If the status flag indicates that this condition is occurring, according to the “YES” path exiting decision block 2340, the fault condition will be identified as an amplifier short circuit, as shown in block 2345. In particular, the fault isolation process will identify the excitation transistor that causes the fault condition, eg, the excitation transistor T1 in the above example.
However, if the status flag does not indicate that this condition has occurred, according to the “NO” path exiting decision block 2340, the status flag will instead have two excitation transistors as indicated by decision block 2350. It may indicate that SHUTDOWN has occurred when the pair is activated and that the same two phase exciters have been engaged in both cases. For example, the status flag may indicate that SHUTDOWN occurred during activation of the transistor pair including the excitation transistors T1 and T5 and that SHUTDOWN occurred during activation of the transistor pair including the excitation transistors T2 and T4. . In this example, only phase exciters A and B are affected. If the status flag indicates that this condition is occurring, the fault will be identified as a motor winding short, as indicated by block 2355, following the “YES” path exiting decision block 2350. . In particular, the fault isolation process is similar to the motor winding M in the above example. _A -M _B Thus, the motor winding that causes the fault condition will be identified.
If the status flag does not indicate that the identified condition exists in decision block 2350, the status flag will have four excitations, as indicated in decision block 2360, according to the “NO” path exiting decision block 2350. It may indicate that SHUTDWON has occurred when the transistor pair is activated, and where all four affected excitation transistor pairs are accompanied by a common exciter. For example, the status flag may indicate that SHUTDOWN has occurred during activation of the transistor pair including the excitation transistors T1 and T5, T1 and T6, T2 and T4, and T3 and T4. In this example, both the upper excitation transistor T1 and the lower excitation transistor T4 related to the phase exciter A are engaged. If the status flag indicates that this condition has occurred, according to the “YES” path from decision block 2360, the fault condition will be identified as a motor winding ground fault, as indicated by block 2365. Become. In particular, the fault isolation process will identify phase exciters that cause fault conditions, such as phase exciter A in the above example.
If the status flag does not indicate that the identified condition exists in decision block 2360, the status flag will follow the “NO” path from decision block 2360, as shown in decision block 2370. It may indicate that a NO CURRENT condition has occurred during the activation of the pair, and where two affected excitation transistor pairs are associated with the same upper or lower excitation transistor. For example, the status flag may indicate that NO CURRENT occurred during activation of the transistor pair including excitation transistors T1 and T5 and activation of the transistor pair including excitation transistors T1 and T6. In this example, both affected excitation transistor pairs are associated with the upper excitation transistor T1. If the status flag indicates that this condition has occurred, the fault condition will be identified as an amplifier open, as indicated by block 2375, following the “YES” path from decision block 2370. . In particular, the fault isolation process can identify an excitation transistor that causes a fault condition, such as the excitation transistor T1 in the above example.
If the status flag does not indicate that the identified condition exists in decision block 2370, the status flag will indicate that the NO CURRENT condition during activation of the four excitation transistor pairs follows the “NO” path from decision block 2370. All four excitation transistor pairs are associated with a common phase exciter. For example, the status flag may indicate that a NO CURRENT condition has occurred during activation of the excitation transistor pairs T1 and T5, T1 and T6, T2 and T4, and T3 and T4. In this example, all of the phase exciters A are common to four excitation transistor pairs. If the status flag indicates that this condition has occurred, the fault condition will be identified as an open motor winding, as indicated by block 2385, following the “YES” path from decision block 2380. . In particular, the fault isolation process is similar to the motor winding M in the above example. _A Motor windings that are causing fault conditions, such as
If the status flag does not indicate that the identified condition exists in decision block 2380, then the fault isolation process will follow the “NO” path exiting decision block 2380, as indicated by block 2390, and a single point fault will occur. It will be displayed that the conditions are not separated.

Table II summarizes the three phase motor, PWM amplifier fault isolation process according to the preferred embodiment of the present invention, where “SD” means SHUTDOWN and “NC” means NO CURRENT condition. What is important is that the three-phase motor, PWM amplifier servo system circuit 2200 is wired in a “WYE” configuration. However, as those skilled in the art will readily appreciate, a similar fault isolation process can be implemented for a “DELTA” wired, three-phase motor, PWM amplifier servo control circuit, and also the spirit of the present invention. It is considered to be included.
FIG. 24 is a circuit diagram of a single phase motor, linear amplifier servo system circuit 2400. Under normal operating conditions, the single-phase motor, linear amplifier servo system circuit 2400 controls the commutation of the current through the single-phase winding, and this current is _S Control the rotation of the axis.
In the preferred embodiment of the present invention, the single phase motor, linear amplifier servo system circuit 2400 employs a single supply, linear transconductance bridge amplifier configuration including a monolithic two power operational amplifier. In particular, this configuration is a power supply V _L Including V here _L Is equal to +28 volts in the preferred embodiment. Power supply V _L Is bidirectional load flow I _LOAD Single phase motor M _S Of the load current I _LOAD Is linearly proportional to the amount of voltage applied to the inputs of operational amplifier A and operational amplifier B.
The single phase motor, linear amplifier servo system circuit 2400 works as follows. When the output voltage of amplifier A and the power supply voltage are equal (eg, equal to +28 volts) and the output voltage of amplifier B is equal to 0 volts, the load current I _LOAD Is at its maximum positive value. When the output voltage of the amplifier B is equal to the power supply voltage and the output voltage of the amplifier A is equal to 0 volts, the load current I _LOAD Is at its maximum negative value. When the output voltages of amplifier A and amplifier B are equal (for example, the outputs of amplifier A and amplifier B are both V _L / When equal to 2 volts) Load current I _LOAD It is logically derived that is zero. In addition, single-phase motor M _S The voltage applied to is 0 volts and + V _L Can vary between volts and therefore the load current I _LOAD Varies linearly with respect to this voltage. Therefore, as will be readily appreciated by those skilled in the art, the single phase motor, linear amplifier servo system circuit 2400 has a magnitude of voltage command (ie, the amount of voltage applied to the inputs of operational amplifier A and operational amplifier B). ) To change servo motor M _S The amount of torque applied to the shaft can be varied linearly.
In FIG. 24, each of linear amplifiers A and B includes an upper excitation transistor (not shown) and a lower excitation transistor (not shown), which are in addition to the upper and lower excitation transistors shown in FIG. Very similar. Therefore, the upper excitation transistor in the linear amplifier A is a single phase motor M. _S Load current I through _LOAD The lower excitation transistor in the linear amplifier B is responsible for generating the positive load current I. _LOAD Sink. Similarly, the upper excitation transistor in the linear amplifier B is a single-phase motor M. _S Negative load current I _LOAD The lower excitation transistor in the linear amplifier A is responsible for generating the negative load current I. _LOAD Sink.
Unlike the PWM amplifier servo assembly control circuits 2000 and 2200, the single phase motor, linear amplifier servo system circuit 2400 has a power supply current I _S Voltage V proportional to the amount of current (ie, the amount of current applied to operational amplifiers A and B) _S Analog power supply current sensing amplifier 2410 generating return current ± I _R Amount (ie + I _R Or -I _R Voltage V proportional to _R Return current sensing amplifier 2415 and single-phase servo motor M _S A shunt resistor connected in parallel with the other winding, where the shunt resistor has a relatively large impedance value. These additional components detect the following single point fault conditions: motor winding open, motor winding wire fault, motor winding ground fault, amplifier excitation transistor open, amplifier short, and amplifier output wire bond open And make it possible to separate.
FIG. 25 is a flowchart showing the detailed steps of the single phase motor, linear amplifier servo fault isolation process. According to FIG. 25, the control circuit 2405 starts the procedure by generating a predetermined amount of positive voltage, as indicated by block 2510. This voltage is applied to the adder 2420 and the adder 2425. In the preferred embodiment, the amount of voltage generated is V _L / Volt (ie, the reference voltage for adders 2420 and 2425), so the input voltage of amplifier A is + V _L Less than volt and the input voltage of amplifier B is + V _L Much smaller than bolts. Under normal operating conditions, this is through a single-phase motor winding, as shown in FIG. _LOAD Produce. Next, as indicated by block 2515, the return current + I _R Is measured.
Control circuit 2405 then generates a predetermined amount of negative voltage, as indicated by block 2520. This negative voltage is applied to the adder 2420 and the adder 2425 in the same manner. Amplifier B is now + V _L The input voltage is slightly lower than volt and amplifier A is + V _L Receive a much lower input voltage. Under normal operating conditions, this is a negative load current I through a single-phase motor winding. _LOAD Will result. As shown in block 2525, the return current -I _R Is now measured.
Control circuit 2405 then returns the return current + I as indicated by decision block 2530. _R Magnitude and return current -I _R Are both smaller than a predetermined (ie, expected) amount of return current, where the expected amount of return current I _Exp Is equal to the amount of return current expected when no fault condition exists. If + I _R And -I _R Is the expected amount of return current I together _EXP If so, according to the “YES” path exiting decision block 2530, the fault condition will be identified as a motor winding open, as shown in block 2535. The motor winding open circuit condition is that the current normally flowing through the motor winding is replaced by the motor M _S Forcing through a shunt resistor 2430 having a large resistance (ie, 1400 ohms in the preferred embodiment) compared to a resistive load of _R Magnitude and return current -I _R The size of I _EXP Smaller than. Therefore, the shunt current I _SHUNT Causes a large voltage drop across the shunt resistor 2430. The large voltage drop across the shunt resistor 2430 reduces the voltage drop that would otherwise occur across the input of the analog current sensing amplifier 2415. This is the return current + I _R Magnitude and return current -I _R Is the expected amount of return current I together _E The return current + I _R And the return current −I as well _R Make it smaller.
However, if the return current + I _R Magnitude and return current -I _R The expected return current I _EXP , The control circuit 2405 follows the “NO” path exiting from the decision block 2530, as shown in decision block 2540, the return current + I _R Magnitude and return current -I _R The size of both is I _EXP While determining whether there is motion of the motor shaft or whether there is erroneous motion of the motor shaft. If these conditions are true, following the “YES” path exiting decision block 2545, the fault condition will be identified as a motor winding fault, as shown in block 2545.
However, if the return current + I _R Magnitude and return current -I _R The size of I _EXP If there is a motor shaft motion and / or there is a motor shaft error motion, then following the “NQ” path exiting decision block 2540, control circuit 2405 returns as indicated by decision block 2550. Current + I _R Magnitude and return current -I _R While determining whether the magnitudes of both are zero, the power supply current I _S Is not equal to the amount of return current (ie, I _S Whether or not is zero). If these conditions are true, the fault condition will be identified as a motor winding ground fault, as indicated by block 2555, following the “YES” path from decision block 2550.
However, if the return current + I _R Magnitude and return current -I _R Is not equal to zero and / or the supply current I _S If the magnitude of is equal to zero, following the “NO” path exiting decision block 2550, the control circuit 2405 determines that the return current + I as indicated by decision block 2560. _R The size of I _EXP The return current −I _R Is determined to be equal to zero or the return current -I _R The size of I _EXP Is equal to the magnitude of the current, while the return current + I _R It is determined whether or not the size of is equal to zero. If either of these conditions is true, the fault condition will be identified as an amplifier-excited transistor open, as indicated by block 2565, following the “YES” path from decision block 2560.
However, if the return current + I _R The size of I _EXP Return current -I not equal to _R Is equal to zero, or the return current −I _R The size of I _EXP The return current + I _R Is equal to zero, the control circuit 2405 determines the return current + I as indicated by block 2570. _R Of expected return current I _EXP , While the return current -I _R Is the predetermined maximum amount of return current I _MAX Determine if greater than or return current -I _R The size of I _EXP Equal to one side, one return current + I _R Is the predetermined maximum amount of return current I _MAX It will be judged whether it is larger. If either of these conditions is true, following the “YES” path from decision block 2570, the fault condition will be identified as an amplifier-excited transistor short as indicated by block 2575. In addition, the control circuit 2405 issues a protection amplifier SHUTDOWN command.
However, if the return current + I _R The size of I _EXP And / or return current -I _R The size of I _MAX If greater, or if return current -I _R The size of I _EXP And / or return current + I _R The size of I _MAX If so, following the “NO” path exiting decision block 2570, the control circuit 2405 determines that the return current + I as indicated by decision block 2580. _R Magnitude and return current -I _R Are both equal to zero, and the power supply current I _S Will also be determined if is zero. If these conditions are true, according to the “YES” path exiting decision block 2580, the fault condition will be identified as an amplifier output wire bond open as indicated by block 2585.
Finally, if the return current + I _R And -I _R Is not zero and / or the supply current I _S If is not equal to zero, a single fault condition will not be identified, as indicated by decision block 2590, following the “NO” path exiting decision block 2580. Importantly, similar procedures that change the order in which steps 2530 through 2585 are performed are considered to be within the spirit of the invention.
Table III summarizes the fault isolation function of the single phase motor, linear amplifier servo fault isolation capability.

In the preferred embodiment of the present invention, the three above identified servo system circuits: two phase motor, PWM amplifier servo system circuit 2000, three phase motor, PWM amplifier servo system circuit 2200, and single phase motor. The linear amplifier servo system circuit 2400 is described as a transconductance device. A transconductance device is a current control device in which the load current (ie, the current flowing through the motor winding) is used as a negative feedback signal to control the input voltage command. However, as those skilled in the art will readily appreciate, these three servo system circuits can also be implemented as voltage controllers rather than current controllers, and as such the corresponding fault isolation process described above. Will not affect.
Signal processing
The present invention provides a number of signal processing techniques designed to enhance the quality of FLIR images and the interface between the image display and the eye. By improving the quality of the FLIR image and its display, the present invention can more accurately draw the AOI and the target within the AOI, and draw it in a safer distance. As shown in FIG. 26, this signal processing technique includes a 2D contrast filter 2605, a bilinear interpolation process (BLI) 2610, a dynamic range control filter 2615, a sub-pixel dithering process 2620, and the FLIR to laser described above. It includes a number of image processing functions such as the FLIR focusing technique 2625 used during the boresight process. Further, this signal processing technique includes notch filter function 2630, analog to digital conversion function 2635, pixel or detector element value gain and level correction function 2640, dead cell replacement function 2645, analog to digital converter (ADC) offset pattern removal. It includes a number of image preprocessing functions such as function 2650.
Starting from the image preprocessing function, the notch filter 2605 is an optical filter made of glass having an antireflection film. The notch filter 2605 is designed to remove the noise signal from the IR signal before the IR signal is focused on the FPA 2607. Specifically, the notch filter 2605 is designed to remove noise signals in the intermediate wave region (ie, 4.2 to 5.55 micrometers) caused by atmospheric radiation. Those skilled in the art will recognize this region of frequency as an atmospheric absorption band or CO. ₂ Often referred to as an absorption band. Without the notch filter 2605, these atmospheric noise signals degrade the IR signal and reduce the quality of the IR image.
A notch filter 2605 is provided in the FLIR detector / cooler assembly 427. Notch filters are generally known in the art and are generally fabricated with FPA cold filters.
When the IR image is focused on the FPA 2607, the analog value (ie optical image) of each detector element is digitized. Hereinafter, each detector element value is referred to as a pixel or a pixel value. Digitization of each detector element takes place in one of four analog-to-digital converters (ADS) 2635. However, as those skilled in the art will appreciate, three or fewer or five or more ADCs may be used. In the preferred embodiment of the invention, each detector element is converted to a 12-bit pixel value.
For each frame of image data, a large amount of image preprocessing is performed on 12-bit pixel values. The first image preprocessing technique is a gain and offset correction process 2640. The purpose of the gain and offset correction process 2640 is to remove specific noise components from the optical image by calibrating each pixel. The noise component removed by calibrating each pixel is caused by fluctuations in gain and offset for each detector element. Variations in gain and offset are transferred to the corresponding pixel values in the digitization process described above. To calibrate, apply hot and cold criteria to each detector element, and adjust each pixel's gain and offset factors as necessary to make each pixel meet the hot and cold criteria. To reflect the same value. The process of calibrating each pixel according to hot and cold criteria is well known in the art.
The next image preprocessing technique is a “dead” cell replacement process 2645. The purpose of this process is to maintain a list of “dead” cells (ie, detector elements that do not respond correctly) and replace the pixel values corresponding to each “dead” cell with the best approximation. The best approximation is obtained by averaging the values of the pixels that touch the pixel corresponding to the “dead” cell. To get the best approximation, use only nearby pixels that correspond to the detector elements that are functioning correctly.
The signal processing subsystem applies any of the well-known criteria to determine which detector elements are “dead”. For example, the thermal response and expected response of each detector element is compared. If the actual response is much greater or less than the expected response, the corresponding detector element is probably not functioning correctly. Another criterion that is often used to determine that a detector element is not functioning properly is whether the digital response of the detector element appears stable or fluctuates. A response that is fluctuating, i.e., wobbling, probably indicates that the corresponding detector element is not functioning properly. Yet another criterion is to compare the response of a given detector element with the average value obtained from the responses of all detector elements. If a response is substantially different from the average response, this probably indicates that the corresponding detector element is not functioning properly. Also, if the dynamic range of a given detector element is limited, this probably indicates that the detector element is not functioning properly. As those skilled in the art will appreciate, this list of criteria is not limited to this, and other criteria can be used to identify "dead" detector elements as well. In general, procedures for replacing “dead” cells are well known in the art.
The next signal preprocessing technique is the ADC offset pattern removal function 2650. The FPA 2607 has four output lines as shown in FIG. Each of these output lines connects to a separate ADC 2635. As explained above, the ADC converts the analog voltage level associated with the detector element to a 12-bit pixel value. Further, each of the four ADCs converts the voltage level of every four columns of the FPA 3607, respectively. For example, the first ADC has columns 1, 5, 9,. . . 477 converts the analog voltage associated with the detector elements in 477. The second ADC has columns 2, 6, 10,. . . The analog voltage associated with the detector elements in 478 is converted. The third ADC has columns 3, 7, 11,. . . 479 converts the analog voltage associated with the detector elements in 479. The fourth ADC has columns 4, 8, 12,. . . 480 converts the analog voltage associated with the detector elements in 480. In addition, ADCs are very sensitive and tend to drift over time, especially when the atmospheric temperature changes. However, if any of the ADCs drift, it can drift independently of the other three ADCs. As one ADC drifts relative to the other three, an undesirable offset or bias occurs every four columns of the digital image. This offset or bias is called an ADC offset pattern.
The purpose of the ADC offset pattern removal function 2650 is to remove this offset or bias from the digital image by adjusting the affected pixel values. In the preferred embodiment of the invention, this process of adjusting the affected pixel values is performed as follows. Four histograms 2660, 2665, 2670, and 2675 are generated for each frame of image data. The contents of each histogram are based on pixel values produced by the corresponding one of the four ADCs. Thus, each histogram reflects 120 columns of pixel values, and each column contains 480 pixel values. For each frame of image data, the signal processing software 2680 performs an average pixel value H for one histogram. ₁ , H ₂ , H _Three , H _Four Are calculated in order. Here, the average pixel value is calculated using only the pixel values between the 20th percentile pixel value and the 80th percentile pixel value. Also, for each frame of image data, the signal processing software uses four individual histogram averages H ₁ , H ₂ , H _Three , H _Four All pixel values average H _BAR Calculate H _BAR And the difference H between the average value of each individual histogram _BAR -H ₁ , H _BAR -H ₂ , H _BAR -H _Three , H _BAR -H _Four Is added to or subtracted from the existing corresponding pixel offset factor, as the case may be.
As previously mentioned, the present invention uses a number of image processing functions to improve digital image quality and the interface between the display and the eye. The first of these image processing functions is a 2D sharpening filter 2605. The 2D sharpening filter 2605 is used for edge enhancement (that is, enhancement of high-frequency image data). In general, edge enhancement is performed by low-pass filtering to 12 bits per pixel input image to generate a low-pass image. If a low-frequency image is subtracted from the input image, a high-frequency image is obtained. Next, after adjusting the relative gain of the low-pass image and the high-pass image, the two images are integrated to form an enhanced image.
FIG. 27 is a preferred embodiment of a 2D sharpening filter 2700 that operates as follows. The 2D sharpening filter 2700 performs a low-pass filtering operation 2705 on each pixel in the 480 × 480 pixel input image to generate a low-pass image. The low-frequency image appears somewhat unclear because it contains low-frequency image data. The low-pass filtering operation is a 3 × 3 convolution process that replaces the value of each pixel in the input image with an average pixel value. For any given pixel in the input image, sum the pixel value and each neighboring pixel value, then divide the sum by the number of pixels used to obtain the sum and calculate the average pixel value for that pixel . For all pixels not on the edge outside the input image, 9 pixels are involved in each total operation. That is, one pixel performing the averaging operation and eight pixels adjacent thereto. This averaging operation is repeated for all pixel values in the input image.
The 2D sharpening filter 2700 also generates a high pass image by subtracting each pixel value in the low pass image from each pixel value in the input image. The subtraction is represented by an adder 2710. The high-frequency image obtained as a result of this subtraction includes high-frequency image data from the input image.
Furthermore, the 2D sharpening filter 2700 generates an image contrast measure 2715. To generate the image contrast measure 2715, first the difference between adjacent pixel values along each row of the input image is calculated. The contrast measure 2715 is then obtained by summing all the difference values. For example, a pure white or pure black input image (ie, an input image containing almost only low frequency components) yields a very low contrast measure. Of course, there is always some noise component (generally high frequency noise). However, an input image showing a checkerboard pattern in which every other pixel is alternately white and black (ie, the input image contains a large high frequency component) produces a very large contrast measure.
Next, the 2D sharpening filter 2700 receives the gain level G of the low-pass image 2720. _L Is the gain level G of the high-frequency image 2725 _H Adjust against or vice versa. An input image with a low contrast measure generally exhibits a relatively low signal-to-noise ratio (SNR). To enhance the details of images that are blurred in high frequency noise, the ratio G _L / G _H To increase. This has the effect of increasing the signal gain level and decreasing the noise gain level. Next, the adjusted gain level G _L And G _H Is given to each pixel in the low-pass image and the high-pass image. An input image with a high contrast measure generally exhibits a high SNR. Ratio G to further improve image quality _L / G _H Lower. This has the effect of further enhancing the high frequency signal already present in the input image. Again, the adjusted gain level G _L And G _H Is given to each pixel in the low-pass image and Takagi image. For input images that have a contrast measure between a very high contrast measure limit and a very low contrast measure limit, the ratio G based on the polynomial curve that establishes the relationship between the contrast measure and the gain level. _L / G _H Make adjustments. In a preferred embodiment of the invention, the ratio G _L / G _H The polynomial curve used to adjust the value is realized by a lookup table. However, as those skilled in the art will readily appreciate, a polynomial curve can be easily realized with an equation.
The 2D sharpening filter then adds each pixel in the now adjusted low pass image and the corresponding pixel in the now adjusted high pass image to produce an enhanced image. Addition is represented by an adder 2730.
The second image processing function used in the present invention is bilinear interpolation. Bilinear interpolation is used to shift the image horizontally or vertically. It is also used to rotate the image and give an electronic zoom. Bilinear interpolation is generally well known in the art.
The third image processing function used in the present invention is a dynamic range control function 2615. In conventional systems such as LANTIRN, adaptive nonlinear mapping is used to enhance the IR image. The nonlinear mapping method uses a Gaussian distribution or a flat distribution centered on the center of the scale. However, as already pointed out, many “easy-to-see” intensity distributions emphasize black, which is more sensitive to changes in gray levels than humans, over white. Therefore, flat distribution and Gaussian distribution centered on the center of the scale generally do not produce the most “easy-to-see” image.
In contrast, since the present invention handles 12 bits per pixel image data, the center of the histogram distribution is not the center of the scale as shown in FIG. This invention uses a Rayleigh distribution. Its center moves closer to one end of the histogram dynamic range to emphasize dark intensity and not bright intensity (ie, to reduce saturation).
The present invention then remaps 12 bits per pixel IR image (4096 quantized values) to 8 bits per pixel image (256 quantized values) to provide a standard 8 bits per pixel. Adapt to the RS-170 video display device. To avoid a substantial reduction in resolution during the remapping process, the present invention remaps a 12-bit image to an 8-bit image non-uniformly using a look-up table. As shown in FIG. 29, the present invention maintains a 1: 1 mapping method and a high level of resolution over a certain region of a 12-bit image distribution, particularly in a dynamic range portion where the density of image data is high. At the same time, the present invention reduces the resolution level of other portions of the dynamic range where the density of the image data is low. This allows the overall image resolution to be maximized despite the pixel display device being limited to 8 bits.
Further, the fourth image processing function is a subpixel dithering function 2620. In the present invention, the FPA 2630 is a 512 × 512 steering array. However, since the standard RS-170 display device can only scan 480 rows of image data, only the 408x480 portion of the steering array is used to generate the IR image. In any case, the FPA 2630 has a windowing mode that can read even smaller portions of the array. For example, a 240 × 240 image can be read out in about ¼ of the time required to read out a 480 × 480 image. The present invention uses the FPA windowing function and FSM 415 to provide a 2 × (2 ×) enhanced image mode. More specifically, to create RS-170 video, the present invention uses FSM 415 to dither the FPA LOS by ½ of the distance between the center of the detector diametrically and downward, 240 × 240 Based on the image, an enhanced 480 × 480 image is obtained.
In the conventional method of creating RS-170 video data by combining image data, the resolution is generally reduced. For example, one of the conventional methods shown in FIG. 30 is to discard the even-numbered detector rows to create an “odd” video field, and to discard the “odd-numbered” detector rows to create an “even” video field. create. This method reduces the sensitivity of the detector because half of the image data is discarded. Another conventional method shown in FIG. 31 averages the rows to create “even” and “odd” video fields. This method reduces the vertical resolution.
In the present invention, the FPA 2630 uses a 240 × 240 image as shown in FIG. 32, and positions 1a, 2a,. . . , 240a. The first step command is then added to the existing feedback control command for FSM 415. The step command causes the FSM 415 to dither the FOS 2630 LOS to the right by half the distance between the centers of the detectors. As a result, the positions 1b, 2b,. . . , 240b, a new 240x240 image is obtained. This image is integrated into positions 1b, 2b,. . . , 240b and positions 1a, 2a,. . . , 240a are alternately arranged to produce a first 240 (vertical) x 480 (horizontal) image. This first 240 × 480 image represents an “odd” video field.
The second step command causes the FSM 415 to dither the FOS 2630 LOS to the left by half of the distance between the centers of the detectors and by 1/2 of the distance between the centers of the detectors. As a result, the LOS of the FPA is changed to positions 1c, 2c,. . . 240c. After integrating this image, a further step command causes the FSM 415 to dither the LOS of the FPA 2630 again to the right by half the distance between the centers of the detectors. As a result, the LOS of the FPA 2630 is moved to positions 1d, 2d,. . . 240d. This image is integrated and the positions 1d, 2d,. . . , 240d and positions 1c, 2c,. . . , 240c are alternately arranged to produce a second 240 (vertical) x 480 (horizontal) image. This second 240x480 image represents an "even" video field. The “odd” and “even” video fields are then interleaved to yield an enhanced 480 × 480 image of the original 240 × 240 image. As will be readily appreciated by those skilled in the art, this enhanced image is an image that is enhanced to twice the original 240 × 240 window.
Other image processing functions such as the FLIR focusing process support subsystems such as the FLIR to laser boresight described above. Returning to FIG. 11, the boresight focusing screen pattern 1100 includes a number of checkerboard patterns. Three 2085 microradian patterns are associated with the WFOV FLIR focusing process, and four small 700 microradian patterns are associated with the NFOV FLIR focusing process. During the WFOV and NFOV focusing procedure shown in FIG. 12, the IR source 1015 in the BSM illuminates the FPA 2630 with a corresponding checkerboard pattern. Importantly, the width of the check illuminated on the FPA 2630 is smaller than the width of the detector element (ie, the distance between the checkers is less than the distance between the pixels) and is readily understood by those skilled in the art. As such, there must be a random phase relationship between the checkerboard pattern and the pixels. A number of contrast measurements are obtained based on the values of the pixels illuminated by the checkerboard pattern when adjusting the FLIR focusing lens. The position of the FLIR focusing lens that produces the best results corresponds to the peak contrast measurement. The FLIR focusing process will be described in detail below.
FIG. 34 is a flowchart showing the detailed steps of the FLIR focusing process. As noted above, the FLIR focusing process (ie, the NFOV or WFOV FLIR focusing process) begins when the IR source 1015 illuminates the FPA 2630 with a corresponding checker pattern, as shown in block 3405. Next, as shown in block 3410, the pixel values of the pixels illuminated by the checkerboard pattern are recorded. Next, as shown in block 3415, a contrast measurement is calculated by calculating the difference between the maximum recorded pixel value and the minimum recorded pixel value using the pixel values described above. Next, as shown in block 3420, the FLIR focusing lens is incrementally adjusted (ie, converted). If additional contrast measurements are required, the additional pixel values are recorded according to the “yes” path from decision block 3425 and the additional contrast measurements are calculated. However, if additional contrast measurements are not needed, follow the “no” path from decision block 3425 to plot the adjustment data as shown in block 3430 (ie, the FLIR focusing lens position relative to the contrast measurement) and block As shown at 3435, the adjustment data points are connected by the best-fit polynomial curve. Next, as shown in block 3440, the maximum adjustment point is determined (ie, the peak of the polynomial curve). The maximum adjustment point represents the FLIR lens position corresponding to the best FLIR image focus.
This process not only provides a clearer and more visually accurate image, but also facilitates a more accurate boresight process and ultimately provides a more accurate FLIR to laser LOS. Furthermore, as described above, if there is no random phase relationship between the checker pattern and the pixel, the recorded pixel value reflects a fixed alias for the checkerboard pattern and the FLIR focusing lens can be adjusted. Does not affect the contrast level of recorded pixel values.
A preferred embodiment of the present invention has been described. However, it will be apparent to those skilled in the art that the present invention may be practiced in specific forms different from those set forth above without departing from the spirit of the invention. The preferred embodiments described above are merely examples and should not be construed as limiting in any way. The scope of the present invention is given not by the above description but by the scope of the claims, and all changes and equivalents included in the scope of the claims are included in the scope of the claims.

Claims (23)

A forward searching infrared (FLIR) optical subsystem for receiving infrared (IR) energy from a region of interest (AOI) and generating an AOI image;
A laser optical system for generating laser energy for irradiating at least one object of the AOI and receiving laser energy reflected by the at least one object;
Any optical element according to the individual pitch rotation is, as will be shared in common by the FLR optical system and the laser optical system, the FLIR optical system and the laser optical system share a common pitch bearing, further
The optical elements shared by the FLIR optical subsystem and the laser optical system are for optically stabilizing the optical image about the axis of rotation in response to FLIR energy of the field line and pitch angle perturbation. Equipped with de-roll means ,
Target and image system characterized by that. The laser optical system is
A laser transmitter;
The target and image system of claim 1, comprising a laser receiver that receives the laser energy reflected by the at least one object. The target and image system of claim 2, wherein the laser transmitter is an ND: YAG laser transmitter. The laser receiver
A laser area receiver;
The target and image system of claim 2, comprising a laser spot tracker. The target and image system of claim 2, wherein the laser transmitter transmits an intermediate wave IR signal. Forward search infrared (FLIR) optical system for receiving infrared (IR) energy from a region of interest (AOI);
A FLIR optical imager arranged to receive the IR energy from the FLIR optical system and for generating an IR image with the IR energy received from the AOI;
A laser transmitter;
A laser receiver and a laser optical device for directing laser energy from the laser transmitter to a desired target located at the AOI and directing laser energy back from the desired target of the AOI to the laser receiver;
Any optical element according to the individual pitch rotation is, as will be shared in common by the FLR optical system and the laser optical system, the FLIR optical system and the laser optical system share a common pitch bearing, further
The optical elements shared by the FLIR optical system and the laser optical means are derolls for optically stabilizing the optical image about the axis of rotation in response to FLIR energy of the field line and pitch angle perturbation. Means,
Target and image system characterized by that. The optical elements shared in common by the FLIR optical system and the laser optics are pitch / yaw focal,
The target and image system according to claim 6, comprising a fast steering mirror (FSM). The target and image system according to claim 6, wherein the deroll means comprises a deroll prism. The FLIR optical system is
Pitch / yaw focal composed of multiple lenses that serve as a common stop for both laser energy and IR energy;
FLIR field view (FOV) optically coupled to pitch / yaw focal and switchable between FLIR narrow field view (NFOV) and FLIR wide field view (WFOV);
Deroll prism,
The target and image system according to claim 6, comprising a fast steering mirror (FSM). The FLIR FOV system
10. The target and image system according to claim 9, further comprising an optical device for holding a real entrance pupil and a real exit pupil on an intermediate focal plane. The FLIR imager is optically coupled to the FLIR optical system and the FLIR optical imager;
A FLIR focusing device for focusing the IR image;
7. A target and image according to claim 6, comprising a FLIR detector for generating an electronic image of an IR image optically coupled to the FLIR imager via a FLIR imager / detector interface. system. 12. The target and image system of claim 11, wherein the FLIR imager and the FLIR detector each comprise a collimating lens for aiming IR energy passing through the FLIR imager / detector interface. . Steering focal plane arrangement,
A cooling device;
The target and image system according to claim 11, further comprising an atmospheric absorption band and a notch filter. The laser optical device comprises:
Pitch / Year Focal,
A laser focusing optical device for focusing the laser energy optically coupled to the pitch / yaw focal;
A combined optical device optically coupled to the laser focus optics and directing laser energy to the laser receiver;
Deroll prism,
A fast steering mirror (FSM) and coupled to the combined optical device via the deroll prism and the FSM and connected to the racer transmitter for manipulating the laser energy and adjusting the polarization of the laser energy The target and image system according to claim 6, further comprising: The laser receiver
The laser region receiver (LRR);
15. The target and image system of claim 14, comprising a laser spot tracker (LST). The combination optical means comprises:
Means for manipulating laser energy in the combined optical means;
16. The target and image system of claim 15 comprising LST / LRR switch means for directing laser energy to one of the LRR and LST. The means for manipulating laser energy in the combination optical means comprises:
The target and image system of claim 16 comprising a set of risley prisms. LRR / LST switch means
The target and image system according to claim 16, comprising an optical wave plate. A set of optical wave plates for adjusting the polarization of the laser energy to ensure that a sufficient amount of laser energy passes through the deroll prism;
15. A target and image system according to claim 14, comprising a set of Risley prisms for manipulating laser energy. The laser optical means includes
Pitch / Year Focal,
A laser focusing optical device for focusing the laser energy optically coupled to the pitch / yaw focal;
Means for reflecting the laser energy from the laser focus means to the laser receiver;
A deroll prism optically coupled to means for reflecting the laser energy to the laser receiving means;
A fast steering mirror (FSM) optically coupled to the deroll prism;
The target and image system of claim 6, comprising a compensator connected to the FSM and the laser transmitter for manipulating the laser energy and adjusting polarization of the laser energy. . 21. The target and image system of claim 20, wherein the laser receiver comprises a laser area receiver (LRR) / laser area spot tracker (LST). 22. The target and image system of claim 21, wherein the LST is a quad cell receiver and the LRR and LST are combined in a single unit so that the LRR is located in the center of the LST. A set of optical wave plates that adjust the polarization of the laser energy to control the amount of laser energy that passes through the deroll prism;
21. The target and image system of claim 20, comprising a set of Risley prisms for manipulating laser energy.

Images