JP3764901B2 - Image radar device - Google Patents

Description

【００４６】
ただし、ｖはドップラー周波数の軸であるため、回転角速度によって伸縮し、その大きさは１でないことに注意する必要がある。また、Ωチルダは次式で与えられるベクトルで、ΩとＷを合成した回転角速度である。
【００４７】
【数２】

【００４８】
レーダ画像のスクリーンの概念を図７に示す。図７において、ｐ＝（ｕ，ｖ）はスクリーン上に投影された位置ベクトルＰ＝（Ｘ，Ｙ，Ｚ）の点目標の像である。このとき、スクリーン上の画像の座標（ｕ，ｖ）は次式で表される。
【００４９】
【数３】

【００５０】
また、スクリーン上の画像の速度は、式（４）、式（５）を時間で微分して得られる。まずｕ軸については、
【００５１】
【数４】

【００５２】
において、
【００５３】
【数５】

【００５４】
を用いて次式が得られる。
【００５５】
【数６】

【００５６】
また、ｖ軸についても、次のように求められる。
【００５７】
【数７】

【００５８】
レーダ画像とその速度分布は式（４）、式（５）、式（９）、式（１０）の４つの式で与えられるが、式（９）はレンジの変化ｕドットがｖと等価であることを示しているので、独立な方程式は高々３本である。そこで、目標上の位置が既知であるｎ点の反射点（ｎ≧３）を同時に観測した場合、次のように書くことができる。
【００５９】
【数８】

【００６０】
ここで、それぞれの行列は次のように定義した。
【００６１】
【数９】

【００６２】
ただし、ＳとΩチルダ、およびΩチルダドットにおける添字の１、２、３は、それぞれのベクトルにおけるＸ、Ｙ、Ｚ軸の要素である。また、それ以外の変数における添字の数字は、ｎ点の反射点を区別するための序数である。
【００６３】
ここで、式（１１）を解くことにより目標の姿勢Ｓを求めることができる。反射点の数ｎがｎ≧３であれば、よく知られた最小２乗法を用いて解くことができて、Ｓの推定値Ｓバーは次式で与えられる。
【００６４】
【数１０】

【００６５】
すなわち、姿勢推定手段２５ａは、図３の構成によって式（２３）を計算して目標の姿勢Ｓの推定値を求める。
【００６６】
同様に、式（１２）を解くことにより目標の角速度Ωチルダを求めることができる。反射点の数ｎがｎ≧３であれば、よく知られた最小２乗法を用いて解くことができて、Ωチルダの推定値Ωチルダバーは次式で与えられる。
【００６７】
【数１１】

【００６８】
すなわち、角速度推定手段２６ａは、図４の構成によって式（２４）を計算して目標の角速度Ωチルダの推定値を求める。
【００６９】
さらにまた同様に、式（１３）を解くことにより目標の角加速度Ωチルダドットを求めることができる。反射点の数ｎがｎ≧３であれば、よく知られた最小２乗法を用いて解くことができて、Ωチルダドットの推定値Ωチルダドットバーは次式で与えられる。
【００７０】
【数１２】

【００７１】
すなわち、角加速度推定手段２７は、図５の構成によって式（２５）を計算して目標の角速度Ωチルダドットの推定値を求める。
【００７２】
このように、目標上に位置が既知あるいは測定可能な反射点が３つ以上あって、そのレーダ画像と、画像上の輝点の速度ベクトル分布が観測できれば、式（２４）および式（２６）の条件の下で、目標の姿勢と角速度、および角加速度を求めることが可能であることが理解される。
【００７３】
このように本実施の形態の構成によれば、輝点選定手段２４がベクトルＰ＝（Ｘ、Ｙ、Ｚ）およびｐ＝（ｕ、ｖ）を求め、画像比較手段２２がｖドットを求め、姿勢推定手段２５ａが式（２３）に従って目標の姿勢を求め、角速度推定手段２６ａが式（２４）に従って目標の角速度を求め、角加速度推定手段２７が式（２５）に従って目標の角加速度を求めるので、目標の運動を測定するレーダ装置を得ることができる。
【００７４】
実施の形態２．
図８はこの発明の画像レーダ装置の他の例を示すブロック図である。図９は図８の姿勢推定手段２５ｂの詳細な構成を示すブロック図である。図１０は図８の角速度推定手段２６ｂの詳細な構成を示すブロック図である。本実施の形態においては、実施の形態１の構成に対して、姿勢推定手段２５ｂおよび角速度推定手段２６ｂに行列式評価手段３３が追加され、さらに姿勢推定手段２５ｂおよび角速度推定手段２６ｂに行列式評価手段３３から、輝点選定手段２４にそれぞれフィードバックループが追加されている。その他の構成は、実施の形態１と同様である。
【００７５】
実施の形態１と同じように、姿勢推定手段２５ｂは、式（２３）を計算して目標の姿勢を求めるが、式（２３）を計算するためには、行列（Γ^T）Γの逆行列が存在する必要がある。その条件は（Γ^T）Γの行列式がゼロでないことであり、次式で表される。
【００７６】
【数１３】

【００７７】
これは、選択した輝点の位置関係に制限が与えられることを表している。すなわち、姿勢推定手段２５ｂは、図９の構成によって式（２３）を計算して目標の姿勢Ｓの推定値を求めるが、式（２６）の条件を計算して、これを満足しない場合には、フィードバックループを介して輝点選定手段２４に輝点の再選定を指示する。従って、式（２６）の条件を満足する輝点が必ず選択されて、目標の姿勢を安定に求められる効果がある。
【００７８】
すなわち、行列式評価手段３３は、行列（Γ^T）Γの行列式を計算し、その値がゼロであった場合には、輝点選定手段２４に対して、輝点の選定を変えるように指示する。
【００７９】
同様に、角速度推定手段２６ｂは、式（２４）を計算して目標の角速度を求めるが、式（２４）を計算するためには、行列（ΓΛ）^T（ΓΛ）の逆行列が存在する必要がある。その条件は（ΓΛ）^T（ΓΛ）の行列式がゼロでないことであり、次式で表される。
【００８０】
【数１４】

【００８１】
これは、選択した輝点の位置関係と姿勢の組み合わせに制限が与えられることを表している。すなわち、角速度推定手段２６ｂは、図１０の構成によって式（２４）を計算して目標の角速度Ωチルダの推定値を求めるが、式（２７）の条件を計算して、これを満足しない場合には、フィードバックループを介して輝点選定手段２４に輝点の再選定を指示する。従って、式（２７）の条件を満足する輝点が必ず選択されて、目標の角速度を安定に求められる効果がある。
【００８２】
すなわち、行列式評価手段３３は、行列（ΓΛ）^T（ΓΛ）の行列式を計算し、その値がゼロであった場合には、輝点選定手段２４に対して、輝点の選定を変えるように指示する。
【００８３】
さらにまた同様に、角加速度推定手段２７は、式（２５）を計算して目標の角加速度を求めるが、式（２５）を計算するためには、行列（ΓΛ）^T（ΓΛ）の逆行列が存在する必要がある。その条件は式（２７）に同じであって、角速度推定手段２６ｂがすでに評価している。従って角加速度推定手段２７は、式（２７）の評価および輝点選定手段２４へのフィードバックループは必要としない。
【００８４】
このように本実施の形態の構成によれば、姿勢推定手段２５ｂが式（２６）の条件を計算して、これを満足しない場合には、フィードバックループを介して輝点選定手段２４に輝点の再選定を指示し、角速度推定手段２６ｂが式（２７）の条件を計算して、これを満足しない場合には、フィードバックループを介して輝点選定手段２４に輝点の再選定を指示するので、これらの条件式を満足する輝点が必ず選択されて、目標の運動を安定に求められる効果がある。
【００８５】
実施の形態３．
図１１はこの発明の画像レーダ装置の他の例を示すブロック図である。図１１において、４０は目標測距手段（測距手段）、４１は座標変換手段、４２は角速度修正手段、４３は角加速度修正手段である。その他の構成は、実施の形態１と同様である。
【００８６】
次に動作について説明する。実施の形態１で説明したように、姿勢推定手段２５ａからは目標の姿勢Ｓが、角速度推定手段２６ａからは目標の角速度Ωチルダが、角加速度推定手段２７からは目標の角加速度Ωチルダドットが、それぞれ得られる。
【００８７】
しかし、式（３）に定義したように、Ωチルダは目標の回転角速度Ωと並進速度Ｗの合成で、ＬＯＳを基準とした角速度である。そこで、目標の回転角速度Ωを求めるためには、式（３）から導かれる次の式を計算すれば良い。
【００８８】
【数１５】

【００８９】
そこで、目標追尾手段７が目標の並進速度Ｗを測定し、目標測距手段４０が目標までの距離ｒ₀を測定し、角速度修正手段４２が式（２８）によって目標の角速度Ωを算出する。
【００９０】
ただし、目標追尾手段７が測定できる目標の並進速度は、図６に示すレーダに固有の座標系ξηζによる値Ｗ’である。一方、式（２８）において必要な目標の並進速度Ｗは目標に固有の座標系ＸＹＺにより定義されているので、座標系の変換が必要である。一般に原点の等しい２つの直交座標系の回転による変換は次式で与えられ、独立な変数はθ，φ，ψの３つである。
【００９１】
【数１６】

【００９２】
ただし、θ，φ，ψは一般にオイラー角と呼ばれる角度で、その関係は図１２に示される。また、目標追尾手段７はアンテナから見た目標の方位を測定できるので、図６のベクトルＳのξηζ座標系における要素を知ることができる。そこで、座標変換手段４１は、式（２９）の右辺のξηζに目標追尾手段７が出力するＳの要素を代入し、左辺のＸＹＺに姿勢推定手段２５ａが出力するＳの要素を代入して、３次の連立方程式を解くことにより３つの変数θ，φ，ψを求める。
続いて、座標変換手段４１は、式（２９）の右辺のξηζに目標追尾手段７が出力する目標の並進速度Ｗ’の要素を代入して、左辺に並進速度Ｗを得る。そこで、角速度修正手段４２が式（２８）によって目標の角速度Ωを算出することができる。
【００９３】
同様にして、目標の回転角加速度Ωドットを求めることができる。式（２８）を時間で微分して次式を得る。
【００９４】
【数１７】

【００９５】
そこで、目標追尾手段７が目標の並進速度Ｗ’を測定し、座標変換手段４１が並進速度Ｗに変換し、目標測距手段４０が目標までの距離ｒ₀を測定し、遅延手段２１ｂと行列加算手段３６ａが目標の姿勢Ｓの時間微分Ｓドットを求め、遅延手段２１ｃと行列加算手段３６ｂが目標の並進速度Ｗの時間微分Ｗドットを求め、角加速度修正手段４３が式（３０）によって目標の角加速度Ωドットを算出する。
【００９６】
このように本実施の形態の構成によれば、目標の回転角速度Ωと、目標の回転角加速度Ωドットとを求めることができて、実施の形態１の発明の効果に加えて、並進運動成分を含まない目標の運動と、並進運動とを分離して求められる効果を有するレーダ装置を得ることができる。
【００９７】
実施の形態４．
図１３はこの発明の画像レーダ装置の他の例を示す輝点選定手段２４および形状データベース２３の構成をブロック図である。図１３において、４４は形状データベース２３に増設されたＲＣＳデータベース、４５はＲＣＳ比較手段である。また、図１４はこの実施の形態の画像レーダ装置が観測する目標の模式図である。図１４において、４６ａ，４６ｂ，４６ｃは目標上の３つの反射点（電波散乱体）で、そのＲＣＳは図示した球の大きさにしたがって異なるものとする。
【００９８】
これらの反射点の座標とそのＲＣＳは、それぞれ形状データベース２３およびＲＣＳデータベース４４に格納されている。そのためには、予め目標のいくつかの部分について座標とＲＣＳを測定し形状データベース２３に格納しておいても良いし、あるいはＲＣＳを都合良く設計したレーダレフレクタを目標上に設置してその座標を記録して形状データベース２３に格納しておいても良い。
【００９９】
次に動作について説明する。図１３に示す装置において、ＲＣＳ比較手段４５は、入力されたレーダ画像の各反射点の輝度すなわちＲＣＳと、ＲＣＳデータベース４４に記録されたＲＣＳを比較して、一致したものについて、その輝点のレーダ画像上の座標（ｕ，ｖ）と、反射点の目標上の座標（Ｘ，Ｙ，Ｚ）を出力する。
【０１００】
すなわち、輝点選定手段２４は、記録されたＲＣＳを手がかりにして輝点を選定し、さらにレーダ画像上の輝点と目標上の反射点を対応付けて、ベクトルＰ＝（Ｘ，Ｙ，Ｚ）およびｐ＝（ｕ，ｖ）を求める。
【０１０１】
このように本実施の形態の構成によれば、輝点選定手段２４は、記録されたＲＣＳを手がかりにして輝点を選定するので、実施の形態１の発明の効果に加えて、輝点の選定がより確実になって、目標の運動をより安定に求められる効果を有するレーダ装置を得ることができる。
【０１０２】
実施の形態５．
図１５はこの発明の画像レーダ装置の他の例を示す輝点選定手段２４および形状データベース２３の構成をブロック図である。図１５において、４７は形状データベース２３に増設された反射点数データベース、４８は反射点数比較手段である。また、図１６はこの実施の形態の画像レーダ装置が観測する目標の模式図である。図１６において、４９ａ，４９ｂ，４９ｃ，４９ｄ，４９ｅ，４９ｆは目標上の密集した複数の反射点である。
【０１０３】
実施の形態４の図１３に示す装置においては、輝点選定手段２４は、記録されたＲＣＳを手がかりにして輝点を選定したが、本実施の形態の図１５の装置においては、密集した反射点の数を手がかりにして輝点を選定する。
【０１０４】
そこで、これらの反射点の座標とその数は、それぞれ形状データベース２３および反射点数データベース４７に格納されているものとする。そのためには、あらかじめ目標の３箇所以上の部分について座標と反射点数を測定し形状データベース２３に格納しておいても良いし、あるいはレーダレフレクタを目標の３箇所以上の部分についてそれぞれ複数設置してその座標と数を記録し形状データベース２３に格納しておいても良い。
【０１０５】
次に動作について説明する。図１５に示す装置において、反射点数比較手段４７は入力されたレーダ画像上で密集して現れる反射点の数と、反射点数データベース４７に記録された反射点数を比較して、一致したものについて、その輝点のレーダ画像上の座標（ｕ，ｖ）と、反射点の目標上の座標（Ｘ，Ｙ，Ｚ）を出力する。
【０１０６】
すなわち、輝点選定手段２４は、記録された反射点数を手がかりにして輝点を選定し、さらにレーダ画像上の輝点と目標上の反射点を対応付けて、ベクトルＰ＝（Ｘ，Ｙ，Ｚ）およびｐ＝（ｕ，ｖ）を求める。
【０１０７】
このように本実施の形態の構成によれば、輝点選定手段２４は、記録された反射点数を手がかりにして輝点を選定するので、実施の形態１の発明の効果に加えて、輝点の選定がより確実になって、目標の運動をより安定に求められる効果を有するレーダ装置を得ることができる。
【０１０８】
尚、反射点の数だけでなく、その分布パターンを手がかりにして輝点を選定することもできる。
【０１０９】
【発明の効果】
この発明に係る画像レーダ装置は、目標を観測してそのレーダ画像を得るレーダ装置において、目標のレーダ画像を連続的に得る画像連続取得手段と、連続する２枚のレーダ画像を比較して、レーダ画像上の輝点の位置変化に基づき輝点の速度分布を得る速度分布取得手段と、予め複数の目標の形状が格納されている形状データベースと、レーダ画像上の輝点を形状データベースの目標の形状と対応付ける対応手段と、レーダ画像、速度分布および目標の形状から目標の運動を求める運動推定手段とを備えている。そのため、目標の運動を測定する画像レーダ装置を得ることができる。
【０１１０】
また、運動推定手段から対応手段へのフィードバックループをさらに備え、対応手段は、運動推定手段の出力が所定の条件を満たさない場合、目標の形状と対応付けを再度行う。そのため、所定の条件を満足する輝点が必ず選択されて、目標の運動を安定に求められる効果がある。
【０１１１】
また、運動推定手段は、レーダ画像上の輝点のレンジ方向の位置関係に基づいて目標の姿勢を求める姿勢推定手段を有する。そのため、目標の姿勢を求める画像レーダ装置を得ることができる。
【０１１２】
また、運動推定手段は、レーダ画像上の輝点のクロスレンジ方向の位置関係に基づいて目標の回転角速度を求める角速度推定手段を有する。そのため、目標の回転角速度を求める画像レーダ装置を得ることができる。
【０１１３】
また、目標の移動速度を求める追尾手段と、目標までの距離を測る測距手段と、目標の移動速度および目標までの距離に基づいて、目標の回転角速度を補正する角速度修正手段とをさらに備えている。そのため、目標の回転角速度を精度よく求める画像レーダ装置を得ることができる。
【０１１４】
また、運動推定手段は、レーダ画像上の輝点のクロスレンジ方向の変化速度に基づいて目標の回転角加速度を求める角加速度推定手段を有する。そのため、目標の回転角加速度を求める画像レーダ装置を得ることができる。
【０１１５】
また、目標の移動速度を求める追尾手段と、目標までの距離を測る測距手段と、移動速度および目標までの距離に基づいて、目標の回転角加速度を補正する角加速度修正手段とをさらに備えている。そのため、目標の回転角加速度を精度よく求める画像レーダ装置を得ることができる。
【０１１６】
また、形状データベースは、各目標が有する異なる電波散乱断面積の少なくとも３つの電波散乱体を目標の形状として格納し、対応手段は、レーダ画像上の少なくとも３つの輝点を形状データベースの目標の形状と対応付ける。そのため、対応付けがより確実になり、目標の運動をより安定に求められる効果を有するレーダ装置を得ることができる。
【０１１７】
また、形状データベースは、各目標が有する密集した複数の電波散乱体を目標の形状として格納し、対応手段は、レーダ画像上の密集した複数の輝点を形状データベースの目標の形状と対応付ける。そのため、対応付けがより確実になり、目標の運動をより安定に求められる効果を有するレーダ装置を得ることができる。
【０１１８】
また、形状データベースは、各目標が有する少なくとも３箇所にそれぞれ異なる数で分布する電波散乱体を目標の形状として格納し、対応手段は、レーダ画像上の少なくとも３箇所にそれぞれ異なる数で分布する輝点を形状データベースの目標の形状と対応付ける。そのため、対応付けがさらに確実になり、目標の運動をより安定に求められる効果を有するレーダ装置を得ることができる。
【０１１９】
さらに、形状データベースは、各目標が有する電波散乱体の分布パターンを目標の形状として格納し、対応手段は、レーダ画像上の輝点の分布パターンを形状データベースの目標の形状と対応付ける。そのため、対応付けがさらに確実になり、目標の運動をより安定に求められる効果を有するレーダ装置を得ることができる。
【図面の簡単な説明】
【図１】 この発明の画像レーダ装置を示すブロック図である。
【図２】 この発明の実施の形態１による画像の時間変化を説明する図である。
【図３】 図１の姿勢推定手段の詳細な構成を示すブロック図である。
【図４】 図１の角速度推定手段の詳細な構成を示すブロック図である。
【図５】 図１の角加速度推定手段の詳細な構成を示すブロック図である。
【図６】 この発明の実施の形態１による観測座標系を説明するレーダの送受信アンテナと目標の位置関係を示す図である。
【図７】 この発明の実施の形態１による観測座標系を説明する補助的な図である。
【図８】 この発明の画像レーダ装置の他の例を示すブロック図である。
【図９】 図８の姿勢推定手段の詳細な構成を示すブロック図である。
【図１０】 図８の角速度推定手段の詳細な構成を示すブロック図である。
【図１１】 この発明の画像レーダ装置の他の例を示すブロック図である。
【図１２】 原点の等しい２つの直交座標系の回転による変換に用いられるオイラー角θ，φ，ψの関係を示す図である。
【図１３】 この発明の画像レーダ装置の他の例を示す輝点選定手段および形状データベースの構成をブロック図である。
【図１４】 画像レーダ装置が観測する目標の模式図である。
【図１５】 この発明の画像レーダ装置の他の例を示す輝点選定手段および形状データベースの構成をブロック図である。
【図１６】 画像レーダ装置が観測する目標の模式図である。
【図１７】 従来のレーダ装置のブロック図である。
【図１８】 従来のレーダ装置による観測のジオメトリを説明するための図である。
【図１９】 従来のレーダ装置による観測のジオメトリを説明するための図である。
【図２０】 画像再生手段の詳細な構成を示すブロック図である。
【符号の説明】
７ 目標追尾手段（追尾手段）、２１ 遅延手段（画像連続取得手段）、２２画像比較手段（速度分布取得手段）、２３ 形状データベース、２４ 輝点選定手段（対応手段）、２５，２５ａ，２５ｂ 姿勢推定手段（運動推定手段）、２６，２６ａ，２６ｂ 角速度推定手段（運動推定手段）、２７ 角加速度推定手段（運動推定手段）、４０ 目標測距手段（測距手段）、４２ 角速度修正手段、４３ 角加速度修正手段。[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an image radar apparatus that obtains a radar image by observing a distant target, and more particularly to an image radar apparatus that can obtain the motion of a target.
[0002]
[Prior art]
FIG. 17 is a block diagram of a conventional radar device described in, for example, Japanese Patent No. 2738244. In FIG. 17, 1 is a transmitter, 2 is a receiver, 3 is a transmission / reception switch, 4 is a transmission / reception antenna, 5 is an image reproduction means, 6 is a correlation calculation means, 7 is a target tracking means, and 8 is a point image response estimation means. , 9 is a target aspect angle estimation unit, 10 is an RCS calculation unit, 11 is a target shape data storage unit, 12 is a convolution integration unit, 13 is a maximum value detection unit, and 14 is a target identification result output unit.
[0003]
FIGS. 18 and 19 are diagrams for explaining the observation geometry by this conventional radar apparatus. In the figure, 4 is a transmission / reception antenna, and 19 is a target.
[0004]
FIG. 20 is a block diagram showing a detailed configuration of the image reproducing means 5 of FIG. In FIG. 20, 15 is a range compression means, 16 is a motion compensation means, 17 is a cross range compression means, and 18 is a two-dimensional storage means.
[0005]
Next, the operation will be described. The high frequency signal generated by the transmitter 1 is radiated from the transmission / reception antenna 4 toward the target 19 via the transmission / reception switch 3. A part of the high frequency signal irradiated to the target 19 is reflected by the target 19 and received again by the transmission / reception antenna 4, amplified and detected by the receiver 2 through the transmission / reception switch 3, and then the target 19 by the image reproducing means 5. Is converted into a radar image representing a RCS (Radar Cross Section) distribution. Hereinafter, a method for reproducing an image will be described in detail.
[0006]
The received signal output from the receiver 2 is input to the image reproducing means 5, and first, the range compression means 15 performs processing for improving the range resolution, that is, pulse compression. The received signal after the range compression is stored in the two-dimensional storage means 18 according to the range bin number m and the pulse hit number n. In order to remove random components harmful to image reproduction from the movement of the target 19, the received signal is read from the two-dimensional storage means 18, and the motion compensation means is set so that the Doppler frequency at the center point 20 of the target 19 becomes zero. The phase compensation and the rearrangement of the range bins are performed by 16 and stored again in the two-dimensional storage means 18.
[0007]
Assuming that the target 19 is rotated by a turning motion as shown in FIG. 18, different points on the target existing in the same range bin generate reflected waves having different Doppler frequencies. Similarly, even when the target 19 is moving straight, different points on the target existing in the same range bin generate reflected waves having different Doppler frequencies. This can be understood from the fact that the motion shown in FIG. 19A is equivalent to the motion shown in FIG.
[0008]
By utilizing this, the cross range compression means 17 performs FFT (Fast Fourier Transform) for each range bin on the signal after the phase compensation, thereby improving the resolution of the cross range which is the direction orthogonal to the range. Through these processes, the received signal is increased in resolution in both the range and cross range directions, and converted into a radar image representing the RCS distribution of each point on the target.
[0009]
On the other hand, the means from the target tracking means 7 to the convolution integration means 12 generate pseudo radar images of various targets under given observation conditions. Although not described in detail because it is not directly related to the present invention, the target tracking means 7 calculates the speed of the target 19, the point image response estimating means 8 calculates the point image response of the radar image, and the target aspect angle estimating means 9 Estimate the target aspect angle. The RCS calculation means 10 calculates the target RCS distribution using the target three-dimensional shape stored in the target shape data storage means 11, and the convolution integration means 12 performs convolution integration of the point image response on the RCS distribution, and the comparison Generate a pseudo radar image for verification.
[0010]
Correlation calculation means 6 obtains a correlation value between the radar image and the sequentially generated pseudo radar image, and maximum value detection means 13 detects the maximum one of the sequentially calculated correlation values, The identification result output unit 14 reads out target information corresponding to the dictionary image having the maximum correlation value, for example, a shape and a name from the target shape data storage unit 11 and outputs the information to output the target automatically by the radar. Realize identification.
[0011]
[Problems to be solved by the invention]
Since the conventional radar apparatus is configured as described above, the target identification can be realized, but the target motion cannot be known.
[0012]
The present invention has been made to solve the above-described problems, and an object thereof is to obtain an image radar apparatus capable of obtaining a target motion.
[0013]
[Means for Solving the Problems]
An image radar apparatus according to the present invention, in a radar apparatus that obtains a radar image by observing a target, compares continuous image acquisition means for continuously acquiring a target radar image with two consecutive radar images, A velocity distribution acquisition means for obtaining a velocity distribution of a bright spot based on a position change of a bright spot on a radar image, a shape database in which a plurality of target shapes are stored in advance, and a bright spot on a radar image as a target of the shape database And means for associating with the shape of the object, and motion estimation means for obtaining the motion of the target from the radar image, the velocity distribution and the shape of the target.
[0014]
In addition, a feedback loop from the motion estimation unit to the response unit is further provided, and the response unit performs association with the target shape again when the output of the motion estimation unit does not satisfy the predetermined condition.
[0015]
Further, the motion estimation means has posture estimation means for obtaining a target posture based on the positional relationship of the bright spots on the radar image in the range direction.
[0016]
The motion estimation means has angular velocity estimation means for obtaining a target rotational angular velocity based on the positional relationship of the bright spots on the radar image in the cross range direction.
[0017]
Further, the apparatus further comprises tracking means for obtaining the target moving speed, distance measuring means for measuring the distance to the target, and angular speed correcting means for correcting the rotational angular speed of the target based on the moving speed of the target and the distance to the target. ing.
[0018]
Further, the motion estimation means includes angular acceleration estimation means for obtaining a target rotational angular acceleration based on the change speed of the bright spot on the radar image in the cross range direction.
[0019]
Further, the apparatus further comprises tracking means for obtaining the target moving speed, distance measuring means for measuring the distance to the target, and angular acceleration correcting means for correcting the rotational angular acceleration of the target based on the moving speed and the distance to the target. ing.
[0020]
In addition, the shape database stores at least three radio wave scatterers having different radio wave scattering cross-sectional areas of each target as the target shape, and the corresponding means uses at least three bright spots on the radar image as the target shape of the shape database. Correlate with.
[0021]
The shape database stores a plurality of dense radio wave scatterers of each target as a target shape, and the correspondence means associates the plurality of dense luminescent spots on the radar image with the target shape of the shape database.
[0022]
In addition, the shape database stores radio wave scatterers distributed in different numbers in at least three locations of each target as target shapes, and the corresponding means has brightnesses distributed in different numbers in at least three locations on the radar image. Associate the points with the target shape in the shape database.
[0023]
Furthermore, the shape database stores the distribution pattern of the radio wave scatterer of each target as the target shape, and the correspondence means associates the distribution pattern of the bright spots on the radar image with the target shape of the shape database.
[0024]
DETAILED DESCRIPTION OF THE INVENTION
Embodiment 1 FIG.
FIG. 1 is a block diagram showing an image radar apparatus of the present invention. In FIG. 1, 1 is a transmitter, 2 is a receiver, 3 is a transmission / reception switch, 4 is a transmission / reception antenna, and 5 is an image reproduction means. These are the same as or equivalent to the conventional radar apparatus shown in FIG.
[0025]
Also, 21 is an image delay means as an image continuous acquisition means for continuously obtaining target radar images, 22 is an image comparison means as a speed distribution acquisition means for obtaining a velocity distribution of bright spots, and 23 is a plurality of target targets in advance. A shape database in which shapes are stored, 24 is a bright spot selection means as a means for associating a bright spot on the radar image with a target shape in the shape database 23, 25a is a posture estimation means for obtaining a target posture, and 26a is a target. An angular velocity estimating means 27 for obtaining the angular velocity of the target, and 27 is an angular acceleration estimating means for obtaining the target angular acceleration. The posture estimation means 25a, the angular velocity estimation means 26a, and the angular acceleration estimation means 27 constitute a motion estimation means for obtaining a target motion.
[0026]
FIG. 2 is a diagram for explaining the temporal change of an image according to the first embodiment of the present invention. FIG. 2 is an example of a ship radar image, where the horizontal axis u is the range and the vertical axis v is the cross range. Moreover, the density | concentration represents RCS, and it turns out that the two-dimensional image of a ship is shown with the shading. (A) and (b) are two images observed with a slight time difference, and it can be seen that (b) is slightly rotated with respect to (a). Reference numeral 28 denotes a position vector in the image (a) of the portion close to the bow.
[0027]
(C) is a diagram for explaining the velocity vector of the image detected from the difference between the two images (a) and (b), and 29 is the velocity vector of the point of the position vector 28 of the image (a). is there.
[0028]
FIG. 3 is a block diagram showing a detailed configuration of the posture estimation means 25a of FIG. FIG. 4 is a block diagram showing a detailed configuration of the angular velocity estimating means 26a of FIG. FIG. 5 is a block diagram showing a detailed configuration of the angular acceleration estimating means 27 of FIG. In these figures, 30 is a matrix transposing means, 31 is a matrix multiplying means, 32 is an inverse matrix calculating means, 34 is a Λ matrix generating means, 35 is a Q matrix generating means, and 36 is a matrix adding means. Note that the matrix multiplication means 31 multiplies a matrix inputted from above or below from a matrix inputted from the left from behind.
[0029]
6 and 7 are diagrams for explaining an observation coordinate system according to the first embodiment of the present invention. FIG. 6 is a diagram showing the positional relationship between the transmission / reception antenna of the radar and the target. In the figure, 37 is the center point of the target, and 38 is the transmission / reception antenna. X, Y, and Z are coordinate systems based on the target center point 37, Ω1, Ω2, and Ω3 are rotational angular velocities around the X, Y, and Z axes, and W1, W2, and W3 are X, X, and X, respectively. This is the translation speed along the Y and Z axes. Further, S is an azimuth vector for viewing the transmission / reception antenna 38 from the target center point 37, and corresponds to a target posture based on a line connecting the transmission / reception antenna and the target.
[0030]
On the other hand, FIG. 7 is an auxiliary diagram for explaining the observation coordinate system, in which 39 is a virtual screen onto which a radar image is projected. P is the position vector of a certain reflection point on the target, and p corresponds to the position vector 28 on the radar image of this reflection point shown in FIG.
[0031]
Next, the operation will be described. In the apparatus shown in FIG. 1, the transmitter 1, the receiver 2, the transmission / reception switch 3, the transmission / reception antenna 4 and the image reproducing means 5 are the same as those of the conventional apparatus shown in FIG. Play. Since the delay means 21 delays the obtained radar image for a certain time, two radar images having a time difference are obtained at the input / output terminals as shown in FIG.
[0032]
Next, the image comparison unit 22 compares the two images to obtain a velocity vector 29 of the reflection point 28. This method can be realized, for example, by pattern matching, using the shape around the reflection point and the reflection intensity as a clue. Alternatively, a technique such as optical flow known for optical images can be used. A set of velocity vectors thus obtained is a velocity distribution.
[0033]
Further, the bright spot selection means 24 selects three or more bright spots that can correspond to the shape database 23 from the obtained radar image, and the position vector P in the three-dimensional space. _i = (X _i , Y _i , Z _i ) And the vector p in the radar image _i = (U _i , V _i ) Is output. Here, i is an ordinal number introduced to distinguish bright spots.
[0034]
The posture estimation means 25a calculates the target posture S from the position vector of the selected bright spot and the range coordinates in the radar image. The detailed operation will be described with reference to FIG. Where Γ is a matrix composed of the position vector P of the selected bright spot, u is a range coordinate u in the radar image of the selected bright spot _i Is a matrix composed of
[0035]
The matrix transposing means 30 is a transpose matrix Γ of the matrix Γ. ^T Ask for. The matrix multiplication means 31a is Γ ^T And Γ. The inverse matrix calculation means 32 is a matrix (Γ ^T ) Γ inverse matrix ((Γ ^T ) Γ) ^-1 Matrix multiplication means 31b obtains (((Γ ^T ) Γ) ^-1 ) And Γ ^T Further, the matrix multiplication means 31c multiplies u, and as a result, the target posture vector S is output.
[0036]
Next, the angular velocity estimating means 26a calculates a target rotational angular velocity Ω tilde from the position vector of the selected bright spot, the cross-range coordinates in the radar image, and the target attitude vector S. The detailed operation will be described with reference to FIG. v is the cross-range coordinate v in the radar image of the selected bright spot _i Is a matrix composed of
[0037]
The Λ matrix generation unit 34 generates a matrix Λ composed of elements of the target posture vector S. The matrix multiplication means 31a multiplies Γ and Λ.
[0038]
The matrix transposing means 30 is a transposed matrix (ΓΛ) of the matrix ΓΛ. ^T Ask for. The matrix multiplication means 31b is (ΓΛ) ^T And ΓΛ. The inverse matrix calculation means 32 is a matrix (ΓΛ) ^T Inverse matrix of (ΓΛ) ((ΓΛ) ^T (ΓΛ)) ^-1 The matrix multiplication means 31c obtains ((ΓΛ) ^T (ΓΛ)) ^-1 And (ΓΛ) ^T Further, the matrix multiplication means 31d multiplies v, and as a result, the target rotational angular velocity Ω tilde is output.
[0039]
Finally, the angular acceleration estimating means 27 calculates the target rotation from the position vector of the selected bright spot, the moving speed in the cross range direction in the radar image, the target attitude vector S, and the target rotational angular velocity Ω tilde. Calculate the angular acceleration Ω tilde dot. The detailed operation will be described with reference to FIG. The v dot indicates the moving speed v in the cross range direction in the radar image of the selected bright spot. _i It is a matrix composed of dots.
[0040]
The Λ matrix generation unit 34 generates a matrix Λ composed of elements of the target posture vector S.
[0041]
The Q matrix generation means 35 generates a matrix Q composed of elements of the target posture vector S and elements of the target rotational angular velocity Ω tilde. The matrix multiplication means 31e multiplies Γ and Q, and the matrix addition means 36 subtracts the matrix ΓQ from v dots.
[0042]
Matrix multiplication means 31a multiplies Γ and Λ, and matrix transposition means 30 transposes matrix ΓΛ (ΓΛ). ^T Ask for. The matrix multiplication means 31b is (ΓΛ) ^T And ΓΛ. The inverse matrix calculation means 32 is a matrix (ΓΛ) ^T Inverse matrix of (ΓΛ) ((ΓΛ) ^T (ΓΛ)) ^-1 The matrix multiplication means 31c obtains ((ΓΛ) ^T (ΓΛ)) ^-1 And (ΓΛ) ^T Further, the matrix multiplication means 31d multiplies the matrix (v dot−ΓQ), and as a result, the target rotational angular acceleration Ω tilde dot is output.
[0043]
Next, the principle of this operation will be described. First, the velocity vector distribution in the radar image is obtained. In the coordinate system shown in FIG. 6, X, Y, and Z are coordinate systems based on the target, and Ω = (Ω ₁ , Ω ₂ , Ω _Three ) Is the angular velocity of the target rotational motion, W = (W ₁ , W ₂ , W _Three ) Is a velocity representing the parallel movement of the target, and P is a position vector of a point on the target. Distance r from the origin of the XYZ coordinates ₀ Has an antenna, and the unit vector on the LOS (Line of Sight) connecting the target origin and the antenna is S.
[0044]
At this time, the radar image can be handled as an image obtained by projecting the target RCS distribution on a screen having the following vectors u and v as coordinate axes.
[0045]
[Expression 1]

[0046]
However, since v is an axis of the Doppler frequency, it should be noted that it expands and contracts depending on the rotational angular velocity, and its size is not 1. The Ω tilde is a vector given by the following equation, and is a rotational angular velocity obtained by combining Ω and W.
[0047]
[Expression 2]

[0048]
The concept of the radar image screen is shown in FIG. In FIG. 7, p = (u, v) is a point target image of the position vector P = (X, Y, Z) projected on the screen. At this time, the coordinates (u, v) of the image on the screen are expressed by the following equations.
[0049]
[Equation 3]

[0050]
The image speed on the screen is obtained by differentiating the equations (4) and (5) with respect to time. First about the u-axis,
[0051]
[Expression 4]

[0052]
In
[0053]
[Equation 5]

[0054]
The following equation is obtained using
[0055]
[Formula 6]

[0056]
The v-axis is also obtained as follows.
[0057]
[Expression 7]

[0058]
The radar image and its velocity distribution are given by the following four equations (4), (5), (9), and (10). In equation (9), the range change u dot is equivalent to v. As it shows, there are at most three independent equations. Therefore, when n reflection points (n ≧ 3) whose positions on the target are known are observed simultaneously, the following can be written.
[0059]
[Equation 8]

[0060]
Here, each matrix was defined as follows.
[0061]
[Equation 9]

[0062]
However, the subscripts 1, 2, and 3 in S, Ω tilde, and Ω tilde dot are elements of the X, Y, and Z axes in the respective vectors. The subscript numbers in the other variables are ordinal numbers for distinguishing n reflection points.
[0063]
Here, the target posture S can be obtained by solving the equation (11). If the number n of reflection points is n ≧ 3, it can be solved using the well-known least square method, and the estimated value S of S is given by the following equation.
[0064]
[Expression 10]

[0065]
That is, the posture estimation means 25a calculates the equation (23) with the configuration of FIG. 3 to obtain the estimated value of the target posture S.
[0066]
Similarly, the target angular velocity Ω tilde can be obtained by solving equation (12). If the number n of reflection points is n ≧ 3, it can be solved using a well-known least square method, and an estimated value of Ω tilde Ω tilde bar is given by the following equation.
[0067]
[Expression 11]

[0068]
That is, the angular velocity estimating means 26a calculates the equation (24) with the configuration of FIG. 4 to obtain the estimated value of the target angular velocity Ω tilde.
[0069]
Similarly, the target angular acceleration Ω tilde dot can be obtained by solving the equation (13). If the number n of reflection points is n ≧ 3, it can be solved using a well-known least square method, and an estimated value of Ω tilde dot Ω tilde dot bar is given by the following equation.
[0070]
[Expression 12]

[0071]
That is, the angular acceleration estimating means 27 calculates the equation (25) with the configuration of FIG. 5 to obtain the estimated value of the target angular velocity Ω tilde dot.
[0072]
Thus, if there are three or more reflection points whose positions are known or measurable on the target and the radar image and the velocity vector distribution of the bright points on the image can be observed, the equations (24) and (26) are obtained. It is understood that the target posture, angular velocity, and angular acceleration can be obtained under the following conditions.
[0073]
As described above, according to the configuration of the present embodiment, the bright spot selection unit 24 calculates the vectors P = (X, Y, Z) and p = (u, v), the image comparison unit 22 calculates v dots, Since the posture estimation means 25a obtains the target posture according to the equation (23), the angular velocity estimation means 26a obtains the target angular velocity according to the equation (24), and the angular acceleration estimation means 27 obtains the target angular acceleration according to the equation (25). A radar apparatus for measuring the target motion can be obtained.
[0074]
Embodiment 2. FIG.
FIG. 8 is a block diagram showing another example of the image radar apparatus of the present invention. FIG. 9 is a block diagram showing a detailed configuration of the posture estimation means 25b of FIG. FIG. 10 is a block diagram showing a detailed configuration of the angular velocity estimating means 26b of FIG. In the present embodiment, a determinant evaluation means 33 is added to the posture estimation means 25b and the angular velocity estimation means 26b, and a determinant evaluation is further added to the posture estimation means 25b and the angular velocity estimation means 26b, compared to the configuration of the first embodiment. A feedback loop is added from the means 33 to the bright spot selecting means 24. Other configurations are the same as those in the first embodiment.
[0075]
As in the first embodiment, the posture estimation means 25b calculates the equation (23) to obtain the target posture. In order to calculate the equation (23), the matrix (Γ ^T ) There must be an inverse matrix of Γ. The condition is (Γ ^T ) The determinant of Γ is not zero, and is expressed by the following equation.
[0076]
[Formula 13]

[0077]
This represents that the positional relationship of the selected bright spots is limited. That is, the posture estimation means 25b calculates the equation (23) by the configuration of FIG. 9 to obtain the estimated value of the target posture S. If the condition of the equation (26) is calculated and is not satisfied, Then, the bright spot selection means 24 is instructed to re-select bright spots via the feedback loop. Therefore, there is an effect that a bright point that satisfies the condition of Expression (26) is always selected and the target posture can be obtained stably.
[0078]
That is, the determinant evaluation means 33 uses the matrix (Γ ^T ) The determinant of Γ is calculated, and when the value is zero, the bright spot selection means 24 is instructed to change the selection of the bright spot.
[0079]
Similarly, the angular velocity estimation means 26b calculates the equation (24) to obtain the target angular velocity. In order to calculate the equation (24), the matrix (ΓΛ) ^T There needs to be an inverse matrix of (ΓΛ). The condition is (ΓΛ) ^T The determinant of (ΓΛ) is not zero, and is expressed by the following equation.
[0080]
[Expression 14]

[0081]
This represents that the combination of the positional relationship and posture of the selected bright spot is limited. That is, the angular velocity estimating means 26b calculates the equation (24) by the configuration of FIG. 10 to obtain the estimated value of the target angular velocity Ω tilde, but calculates the condition of the equation (27) and does not satisfy this condition. Instructs the bright spot selection means 24 to reselect bright spots via a feedback loop. Therefore, there is an effect that a bright spot satisfying the condition of Expression (27) is always selected and the target angular velocity can be obtained stably.
[0082]
That is, the determinant evaluation means 33 calculates the matrix (ΓΛ) ^T The determinant of (ΓΛ) is calculated, and when the value is zero, the bright spot selection means 24 is instructed to change the bright spot selection.
[0083]
Furthermore, similarly, the angular acceleration estimation means 27 calculates the equation (25) to obtain the target angular acceleration, but in order to calculate the equation (25), the matrix (ΓΛ) ^T There needs to be an inverse matrix of (ΓΛ). The conditions are the same as in equation (27), and the angular velocity estimation means 26b has already evaluated. Therefore, the angular acceleration estimating means 27 does not require the evaluation of the equation (27) and the feedback loop to the bright spot selecting means 24.
[0084]
As described above, according to the configuration of the present embodiment, when the posture estimation unit 25b calculates the condition of the equation (26) and does not satisfy the condition, the bright point selection unit 24 is informed via the feedback loop. When the angular velocity estimation means 26b calculates the condition of the equation (27) and does not satisfy the condition, the bright spot selection means 24 is instructed to reselect the bright spot via the feedback loop. Therefore, a bright point that satisfies these conditional expressions is always selected, and there is an effect that the target motion can be obtained stably.
[0085]
Embodiment 3 FIG.
FIG. 11 is a block diagram showing another example of the image radar apparatus of the present invention. In FIG. 11, reference numeral 40 is a target distance measuring means (ranging means), 41 is a coordinate converting means, 42 is an angular velocity correcting means, and 43 is an angular acceleration correcting means. Other configurations are the same as those in the first embodiment.
[0086]
Next, the operation will be described. As described in the first embodiment, the target posture S from the posture estimation means 25a, the target angular velocity Ω tilde from the angular velocity estimation means 26a, and the target angular acceleration Ω tilde dot from the angular acceleration estimation means 27. Respectively.
[0087]
However, as defined in Equation (3), the Ω tilde is a combination of the target rotational angular velocity Ω and the translational velocity W, and is an angular velocity based on LOS. Therefore, in order to obtain the target rotational angular velocity Ω, the following equation derived from Equation (3) may be calculated.
[0088]
[Expression 15]

[0089]
Therefore, the target tracking means 7 measures the target translation speed W, and the target distance measuring means 40 is the distance r to the target. ₀ Then, the angular velocity correcting means 42 calculates the target angular velocity Ω by the equation (28).
[0090]
However, the target translation speed that can be measured by the target tracking means 7 is a value W ′ based on the coordinate system ξηζ unique to the radar shown in FIG. On the other hand, since the target translation speed W required in the equation (28) is defined by the coordinate system XYZ specific to the target, it is necessary to convert the coordinate system. In general, conversion by rotation of two orthogonal coordinate systems having the same origin is given by the following equation, and there are three independent variables, θ, φ, and ψ.
[0091]
[Expression 16]

[0092]
However, θ, φ, and ψ are angles generally called Euler angles, and the relationship is shown in FIG. Further, since the target tracking means 7 can measure the direction of the target as viewed from the antenna, it can know the elements of the vector S in FIG. 6 in the ξηζ coordinate system. Therefore, the coordinate conversion means 41 substitutes the element of S output by the target tracking means 7 for ξηζ on the right side of the equation (29), and substitutes the element of S output by the posture estimation means 25a for XYZ on the left side, Three variables θ, φ, and ψ are obtained by solving the cubic simultaneous equations.
Subsequently, the coordinate conversion means 41 obtains the translation speed W on the left side by substituting the element of the target translation speed W ′ output by the target tracking means 7 into ξηζ on the right side of Expression (29). Therefore, the angular velocity correcting means 42 can calculate the target angular velocity Ω by the equation (28).
[0093]
Similarly, the target rotational angular acceleration Ω dot can be obtained. The following equation is obtained by differentiating the equation (28) with respect to time.
[0094]
[Expression 17]

[0095]
Therefore, the target tracking means 7 measures the target translation speed W ′, the coordinate conversion means 41 converts it into the translation speed W, and the target distance measurement means 40 is the distance r to the target. ₀ The delay means 21b and the matrix addition means 36a obtain the time differential S dot of the target posture S, the delay means 21c and the matrix addition means 36b obtain the time differential W dot of the target translation speed W, and correct the angular acceleration. The means 43 calculates a target angular acceleration Ω dot by the equation (30).
[0096]
Thus, according to the configuration of the present embodiment, the target rotational angular velocity Ω and the target rotational angular acceleration Ω dot can be obtained. In addition to the effects of the invention of the first embodiment, the translational motion component Thus, a radar apparatus having an effect obtained by separating a target motion that does not include a motion and a translational motion can be obtained.
[0097]
Embodiment 4 FIG.
FIG. 13 is a block diagram showing the configuration of the bright spot selecting means 24 and the shape database 23 showing another example of the image radar apparatus of the present invention. In FIG. 13, 44 is an RCS database added to the shape database 23, and 45 is an RCS comparison means. FIG. 14 is a schematic diagram of a target observed by the image radar apparatus of this embodiment. In FIG. 14, 46a, 46b and 46c are three reflection points (radio wave scatterers) on the target, and their RCSs differ according to the size of the sphere shown.
[0098]
The coordinates of these reflection points and their RCS are stored in the shape database 23 and the RCS database 44, respectively. For this purpose, the coordinates and RCS of some parts of the target may be measured and stored in the shape database 23 in advance, or a radar reflector that is conveniently designed for RCS may be installed on the target and the coordinates may be stored. May be recorded and stored in the shape database 23.
[0099]
Next, the operation will be described. In the apparatus shown in FIG. 13, the RCS comparison means 45 compares the brightness of each reflection point of the input radar image, that is, the RCS with the RCS recorded in the RCS database 44, and finds the bright spot of the matched one. The coordinates (u, v) on the radar image and the coordinates (X, Y, Z) on the target of the reflection point are output.
[0100]
That is, the bright spot selecting means 24 selects a bright spot using the recorded RCS as a clue, and further associates the bright spot on the radar image with the reflective spot on the target, and the vector P = (X, Y, Z ) And p = (u, v).
[0101]
As described above, according to the configuration of the present embodiment, the bright spot selecting means 24 selects a bright spot using the recorded RCS as a clue, so in addition to the effect of the invention of the first embodiment, It is possible to obtain a radar apparatus having an effect that selection becomes more reliable and a desired motion can be obtained more stably.
[0102]
Embodiment 5. FIG.
FIG. 15 is a block diagram showing the configuration of the bright spot selecting means 24 and the shape database 23 showing another example of the image radar apparatus of the present invention. In FIG. 15, 47 is a reflection point number database added to the shape database 23, and 48 is a reflection point number comparison means. FIG. 16 is a schematic diagram of a target observed by the image radar apparatus of this embodiment. In FIG. 16, 49a, 49b, 49c, 49d, 49e, and 49f are a plurality of dense reflection points on the target.
[0103]
In the apparatus shown in FIG. 13 according to the fourth embodiment, the bright spot selecting means 24 selects bright spots using the recorded RCS as a clue. However, in the apparatus shown in FIG. Select bright spots based on the number of points.
[0104]
Therefore, it is assumed that the coordinates of these reflection points and the number thereof are stored in the shape database 23 and the reflection point number database 47, respectively. For this purpose, the coordinates and the number of reflection points may be measured in advance for three or more parts of the target and stored in the shape database 23, or a plurality of radar reflectors may be installed for each of the three or more parts of the target. The coordinates and number may be recorded and stored in the shape database 23.
[0105]
Next, the operation will be described. In the apparatus shown in FIG. 15, the reflection point number comparison means 47 compares the number of reflection points that appear densely on the input radar image with the number of reflection points recorded in the reflection point number database 47. The coordinates (u, v) on the radar image of the bright spot and the coordinates (X, Y, Z) on the target of the reflection point are output.
[0106]
That is, the bright spot selecting means 24 selects a bright spot using the number of recorded reflection points as a clue, and further associates the bright spot on the radar image with the reflective spot on the target, and the vector P = (X, Y, Z) and p = (u, v) are determined.
[0107]
As described above, according to the configuration of the present embodiment, the bright spot selecting means 24 selects the bright spot based on the number of recorded reflection points, so in addition to the effect of the invention of the first embodiment, the bright spot. Therefore, it is possible to obtain a radar apparatus having an effect that the desired motion can be obtained more stably.
[0108]
In addition, not only the number of reflection points but also the distribution pattern can be used as a clue to select bright points.
[0109]
【The invention's effect】
An image radar apparatus according to the present invention, in a radar apparatus that obtains a radar image by observing a target, compares continuous image acquisition means for continuously acquiring a target radar image with two consecutive radar images, A velocity distribution acquisition means for obtaining a velocity distribution of a bright spot based on a position change of a bright spot on a radar image, a shape database in which a plurality of target shapes are stored in advance, and a bright spot on a radar image as a target of the shape database And means for associating with the shape of the object, and motion estimation means for obtaining the motion of the target from the radar image, the velocity distribution and the shape of the target. Therefore, an image radar device that measures the target motion can be obtained.
[0110]
In addition, a feedback loop from the motion estimation unit to the response unit is further provided, and the response unit performs association with the target shape again when the output of the motion estimation unit does not satisfy the predetermined condition. Therefore, there is an effect that a bright spot satisfying a predetermined condition is always selected and a desired motion can be obtained stably.
[0111]
Further, the motion estimation means has posture estimation means for obtaining a target posture based on the positional relationship of the bright spots on the radar image in the range direction. Therefore, an image radar device that obtains the target posture can be obtained.
[0112]
The motion estimation means has angular velocity estimation means for obtaining a target rotational angular velocity based on the positional relationship of the bright spots on the radar image in the cross range direction. Therefore, an image radar device that obtains the target rotational angular velocity can be obtained.
[0113]
Further, the apparatus further comprises tracking means for obtaining the target moving speed, distance measuring means for measuring the distance to the target, and angular speed correcting means for correcting the rotational angular speed of the target based on the moving speed of the target and the distance to the target. ing. Therefore, it is possible to obtain an image radar apparatus that accurately obtains the target rotational angular velocity.
[0114]
Further, the motion estimation means includes angular acceleration estimation means for obtaining a target rotational angular acceleration based on the change speed of the bright spot on the radar image in the cross range direction. Therefore, an image radar device that obtains the target rotational angular acceleration can be obtained.
[0115]
Further, the apparatus further comprises tracking means for obtaining the target moving speed, distance measuring means for measuring the distance to the target, and angular acceleration correcting means for correcting the rotational angular acceleration of the target based on the moving speed and the distance to the target. ing. Therefore, it is possible to obtain an image radar apparatus that accurately obtains the target rotational angular acceleration.
[0116]
In addition, the shape database stores at least three radio wave scatterers having different radio wave scattering cross-sectional areas of each target as the target shape, and the corresponding means uses at least three bright spots on the radar image as the target shape of the shape database. Correlate with. For this reason, it is possible to obtain a radar apparatus having an effect that the association becomes more reliable and the target motion can be obtained more stably.
[0117]
The shape database stores a plurality of dense radio wave scatterers of each target as a target shape, and the correspondence means associates the plurality of dense luminescent spots on the radar image with the target shape of the shape database. For this reason, it is possible to obtain a radar apparatus having an effect that the association becomes more reliable and the target motion can be obtained more stably.
[0118]
In addition, the shape database stores radio wave scatterers distributed in different numbers in at least three locations of each target as target shapes, and the corresponding means has brightnesses distributed in different numbers in at least three locations on the radar image. Associate the points with the target shape in the shape database. For this reason, it is possible to obtain a radar apparatus having an effect that the association is further ensured and the target motion can be obtained more stably.
[0119]
Furthermore, the shape database stores the distribution pattern of the radio wave scatterer of each target as the target shape, and the correspondence means associates the distribution pattern of the bright spots on the radar image with the target shape of the shape database. For this reason, it is possible to obtain a radar apparatus having an effect that the association is further ensured and the target motion can be obtained more stably.
[Brief description of the drawings]
FIG. 1 is a block diagram showing an image radar apparatus of the present invention.
FIG. 2 is a diagram for explaining a temporal change of an image according to the first embodiment of the present invention.
3 is a block diagram showing a detailed configuration of posture estimation means in FIG. 1. FIG.
4 is a block diagram showing a detailed configuration of angular velocity estimation means in FIG. 1; FIG.
5 is a block diagram showing a detailed configuration of angular acceleration estimation means in FIG. 1. FIG.
6 is a diagram showing a positional relationship between a transmission / reception antenna of a radar and a target for explaining an observation coordinate system according to Embodiment 1 of the present invention; FIG.
FIG. 7 is an auxiliary diagram for explaining an observation coordinate system according to the first embodiment of the present invention.
FIG. 8 is a block diagram showing another example of the image radar apparatus of the present invention.
9 is a block diagram showing a detailed configuration of posture estimation means in FIG. 8. FIG.
10 is a block diagram showing a detailed configuration of the angular velocity estimation means of FIG.
FIG. 11 is a block diagram showing another example of the image radar apparatus of the present invention.
FIG. 12 is a diagram illustrating a relationship between Euler angles θ, φ, and ψ used for conversion by rotation of two orthogonal coordinate systems having the same origin.
FIG. 13 is a block diagram showing the configuration of a bright spot selecting means and a shape database showing another example of the image radar apparatus of the present invention.
FIG. 14 is a schematic diagram of a target observed by the image radar apparatus.
FIG. 15 is a block diagram showing the configuration of a bright spot selecting means and a shape database showing another example of the image radar apparatus of the present invention.
FIG. 16 is a schematic diagram of a target observed by the image radar apparatus.
FIG. 17 is a block diagram of a conventional radar device.
FIG. 18 is a diagram for explaining the geometry of observation by a conventional radar apparatus.
FIG. 19 is a diagram for explaining the geometry of observation by a conventional radar apparatus.
FIG. 20 is a block diagram showing a detailed configuration of image reproduction means.
[Explanation of symbols]
7 target tracking means (tracking means), 21 delay means (continuous image acquisition means), 22 image comparison means (velocity distribution acquisition means), 23 shape database, 24 bright spot selection means (corresponding means), 25, 25a, 25b Estimating means (motion estimating means), 26, 26a, 26b Angular velocity estimating means (motion estimating means), 27 Angular acceleration estimating means (motion estimating means), 40 Target ranging means (ranging means), 42 Angular velocity correcting means, 43 Angular acceleration correction means.

Claims (11)

In a radar device that observes a target and obtains its radar image,
Image continuous acquisition means for continuously obtaining the target radar image;
Speed distribution acquisition means for comparing two successive radar images and obtaining a velocity distribution of the bright spot based on a change in position of the bright spot on the radar image;
A shape database in which a plurality of target shapes are stored in advance;
Corresponding means for associating the bright spot on the radar image with the target shape in the shape database;
An image radar apparatus comprising: a motion estimation unit that obtains a target motion from the radar image, the velocity distribution, and a target shape. A feedback loop from the motion estimation means to the response means;
2. The image radar apparatus according to claim 1, wherein the correspondence unit re-associates with the target shape when the output of the motion estimation unit does not satisfy a predetermined condition. The image radar apparatus according to claim 1, wherein the motion estimation unit includes a posture estimation unit that obtains a target posture based on a positional relationship of a bright spot on the radar image in a range direction. The image according to any one of claims 1 to 3, wherein the motion estimation unit includes an angular velocity estimation unit that obtains a target rotational angular velocity based on a positional relationship in a cross range direction of a bright spot on the radar image. Radar device. Tracking means for determining the target moving speed;
Ranging means to measure the distance to the target,
5. The image radar apparatus according to claim 4, further comprising angular velocity correcting means for correcting the rotational angular velocity of the target based on the moving speed of the target and the distance to the target. The said motion estimation means has an angular acceleration estimation means which calculates | requires a target rotational angular acceleration based on the change speed of the cross range direction of the bright spot on the said radar image, The one of Claims 1 thru | or 5 characterized by the above-mentioned. Image radar equipment. Tracking means for determining the target moving speed;
Ranging means to measure the distance to the target,
7. The image radar apparatus according to claim 6, further comprising angular acceleration correcting means for correcting the rotational angular acceleration of the target based on the moving speed and the distance to the target. The shape database stores at least three radio wave scatterers having different radio wave scattering cross sections each target has as the target shape,
8. The image radar apparatus according to claim 1, wherein the correspondence unit associates at least three bright spots on the radar image with the target shape of the shape database. The shape database stores a plurality of dense radio wave scatterers of each target as the shape of the target,
The image radar apparatus according to claim 1, wherein the correspondence unit associates a plurality of dense luminescent spots on the radar image with the target shape of the shape database. The shape database stores radio wave scatterers distributed in different numbers in at least three locations of each target as the shape of the target,
8. The image radar according to claim 1, wherein the correspondence unit associates bright points distributed in different numbers in at least three places on the radar image with the target shape of the shape database. 9. apparatus. The shape database stores the distribution pattern of the electromagnetic wave scatterer that each target has as the shape of the target,
8. The image radar apparatus according to claim 1, wherein the correspondence unit associates a distribution pattern of bright spots on the radar image with the target shape of the shape database.

Images