JP3749323B2 - Mobile device - Google Patents

Description

【００７６】
を予め設定しておき、評価関数から求められた値の大小関係により選択しても良い。
図１１に戻って説明を続ける。ステップＳ３−２で以上処理により移動経路が生成されると、中央演算処理装置７は、移動処理を開始し、ステップＳ３−２で生成された移動経路に従って移動を開始する（ステップＳ３−３）。このとき、中央演算処理装置７は、選択された移動経路のツアーから順に識別番号を読み出し、読み出された識別番号から識別番号に対応したランドマーク情報を読み出し、読み出されたランドマーク情報と環境センサ４により検出され、抽出された物体とを比較し、目的のランドマークか否かを判断を行いつつ、次々にランドマークを追うことにより移動を実現している。
【００７７】
環境センサ４は、例えば、超音波センサや赤外線センサなどからなり、周囲の状況を画像として捕らえる。中央演算処理装置７は、環境センサ４により捕らえた画像をを解析し、画像から物体（ドア、コーナー、部屋）等を認識し、周囲の状況判断を行う（ステップＳ３−４）。なお、環境センタ４での検出する周囲の状況は画像に限ることはなく、要はランドマークを識別できればよく、単純なソナーセンサやレンジファインダ接触センサ等でも良い。すなわち、センサから得られた情報を抽象化し、予めわかているランドマーク（プロトタイプのドア）等のセンサイメージと比較して識別する程度であればよい。
【００７８】
このとき、移動経路に地図情報とは異なる環境変化、例えば、移動すべき経路上に障害物等を発見した場合、このままでは通行不可能であるので、経路を変更するための地図情報更新処理を行う（ステップＳ３−５）。ステップＳ３−５が請求項１４、１５、１７に相当する処理を実行する。
【００７９】
上記ステップＳ３−３、Ｓ３−４、Ｓ３−５を繰り返しながら最終目的地に移動を行う（ステップＳ３−６）。
図１５に本発明の一実施例の地図情報更新処理の動作フローチャートを示す。
まず、中央演算処理装置７は、検出された環境変化を解析して、移動経路に影響があるか否かを判断する（ステップＳ５−１、Ｓ５−２、Ｓ５−３）。例えば、環境センサ４により開口部を検出すると、開口部の大きさと、自装置の大きさとを比較して、通行可能かどうかを判断することにより、実現される。また、環境センサ４により障害物を検出すると、移動経路上であれば、障害物と移動経路と間隔と自装置の大きさとを比較して、移動経路上を通行可能か否かを検出することにより実現される。
【００８０】
ステップＳ５−２で、新たに通行可能な経路が発見された場合には、中央演算処理装置７は、ランドマーク、及び、リンクの追加を行う（ステップＳ５−４、Ｓ５−５）。
図１６、図１７に本発明の一実施例のランドマーク及びリンクの追加動作の動作説明図を示す。図１６（Ａ）はランドマークの配置を示し、図１６（Ｂ）は地図情報のランドマークを識別番号化した地図、図１７（Ａ）は追加前のツアー群、図１７（Ｂ）は追加後のツアー群を示す。
【００８１】
図１６（Ａ）に示すように部屋３１に新たな出入口３２を認識したとすると、中央演算処理装置７は、まず、出入口３２の部屋３２の内側部分に新たなランドマークＤｉを設定し、外側に新たなランドマークＤｏを設定する。次に、図１６（Ｂ）に示すように新たに設定されたランドマークＤｉ、Ｄｏに、予め設定されていた識別番号ＮＯ．１〜ＮＯ．１５に続けて識別番号ＮＯ．１６、ＮＯ．１７を設定するとともに、リンク３３、３４を追加する。また、リンク３３、３４には図８に示すように移動情報として移動距離ａ、移動時間ｂが座標から算出され、付与され、また、存在確率が初期値０．５に設定される。
【００８２】
なお、このとき、識別番号ＮＯ．１６が識別番号ＮＯ．１１とＮＯ．１４とを結ぶリンク上に位置する場合には、リンクを識別番号１１とＮＯ．１６とを結ぶリンク３５と識別番号ＮＯ．１４とＮＯ．１６とを結ぶリンク３６に分割するようにしてもよい。
【００８３】
以上のようにして、識別番号ＮＯ．１６、ＮＯ．１７のランドマーク及びリンクが追加されると、図１０、図１４に示すツアー群のデータが変更される。識別番号ＮＯ．１６、ＮＯ．１７のランドマーク及びリンクが追加される前、図１７（Ａ）に示すようにツアーＥが識別番号ＮＯ．７→ＮＯ．８→ＮＯ．９であったものが、識別番号ＮＯ．１６、ＮＯ．１７のランドマーク及びリンクの追加により図１７（Ｂ）に示すように、ツアーＥで識別番号ＮＯ．７→ＮＯ．８→ＮＯ．９→ＮＯ．１７→ＮＯ．１６となる。また、図１７（Ａ）に示すツアーＨの識別番号ＮＯ．１１→ＮＯ．１４が図１７（Ｂ）に示すツアーＨの識別番号ＮＯ．１１→ＮＯ．１６、及び、ツアーＩの識別番号ＮＯ．１４→ＮＯ．１６が追加される。
【００８４】
また、ステップＳ５−２で、通行不可能な経路が発見された場合には、中央演算処理装置７は、ランドマーク、及び、リンクの削除を行う（ステップＳ５−６、Ｓ５−７）。
図１８、図１９に本発明の一実施例のランドマーク及び削除時の動作説明図を示す。図１８（Ａ）はランドマークの配置を示し、図１８（Ｂ）は地図情報のランドマークを識別番号化した地図、図１９（Ａ）は削除前のツアー群、図１９（Ｂ）は削除後のツアー群を示す。
【００８５】
図１８（Ａ）に示すように移動経路上に障害物が置かれおり、通行不可能な場合、図１８（Ｂ）に示すように識別番号ＮＯ．６のランドマークと識別番号ＮＯ．７のランドマークとの間の通行が不可能となるので、識別番号ＮＯ．６のランドマークと識別番号ＮＯ．７のランドマークとを接続するリンクを切断する。
【００８６】
識別番号ＮＯ．６のランドマークと識別番号ＮＯ．７のランドマークとを接続するリンクが切断されると、追加時と同様に図１０、図１４に示すツアー群の情報が変更される。識別番号ＮＯ．６、ＮＯ．７の間のリンクが削除される前、図１９（Ａ）に示すように設定されていたツアーＤが図１９（Ｂ）に示すように削除される。
【００８７】
以上により、通行可能となった経路に関するランドマーク、リンクの追加、及び、削除が行われ、新たにツアー群の作成が完了する。
図１５に戻って説明を続けると、以上のようにして、ランドマーク、リンクの追加、及び、削除が行われると、通過できない場合があるので、新たに作成されたツアー群をもとに、再び経路計画を作成を行う（ステップＳ５−８）。なお、経路計画作成は、既に図１２で説明しているのでその説明は省略する。
【００８８】
以上にように、移動中に環境センサ４により環境の変化を検出した場合には、検出した環境に応じてランドマーク及びリンクを書き換え、ツアー群の変更して、新たに経路を作成し、移動を続ける。
以上の処理によれば、ランドマーク地図の変更はロボットが新しくいける場所ができた時といけなくなってしまった場所ができた場合に起こり、それぞれ必要があればランドマークの追加・削除を行ない、さらに、リンクの作成・削除を行なってデータベースのツアー群の再構成を行なう。このような変更は、絶対的な位置で示された地図を変更するのに比べ、非常に簡単な処理となる。
【００８９】
また、このような変更は、自律移動ロボット装置１が環境の変化を検出した時、自律移動ロボット装置１の行動に影響が出る場合のみランドマーク地図の変更を行なう。つまり、自律移動ロボット装置１の移動に無関係な細かい環境の変化やちょっとしたセンサの誤差などは無視することができ、中央演算処理装置７での計算量の削滅をできる。
【００９０】
ここで、再び、図１１に戻って説明を続ける。
このとき、中央演算処理装置７は、移動距離、及び、移動時間を計測し、記憶しており、移動終了後、計測しておいた移動距離、及び、移動時間に応じて地図用メモリ３にデータベース化されている移動情報の更新を行い、正確な情報を与える。
【００９１】
ステップＳ３−６で、最終目的地に到着すると、中央演算処理装置７は、ロボット移動処理を終了して、上記保持された情報に応じて移動情報などを変更する情報変更処理を行う（ステップＳ３−７）。このステップＳ３−７が請求項１２、１３、１４、１５、１６の実施例に相当する。
【００９２】
図２０に本発明の一実施例の情報更新処理の動作フローチャートを示す。
情報更新処理は特許請求の範囲の請求項８及び９の実施例に相当する。
情報更新処理では、中央演算処理装置７は、移動中の記録しておいた各ランドマーク間の移動距離、移動時間を取得する（ステップＳ６−１）。次に、中央演算処理装置７は、ステップＳ６−１で取得した移動距離、移動時間、及び、過去の移動時の移動距離、及び、移動時間から最新の移動距離、及び、移動時間、存在確率を算出し、更新する（ステップＳ６−２、Ｓ６−３）。
【００９３】
図２１に本発明の一実施例の最新移動距離及び移動時間の算出方法を説明するための図を示す。
最新の移動距離、及び、移動時間の算出方法としては過去のデータとの平均値をとる方法がる。
【００９４】
例えば、図７の識別番号ＮＯ．１のランドマークと識別番号ＮＯ．２のランドマークとの間で最初の移動結果、今回、移動距離が９９５．３〔ｃｍ〕、移動時間が１９．９１〔ｍｉｎ〕が得られたとする。このとき、今回の移動が３回目とする。このとき、１回目の移動距離が９８５．３〔ｃｍ〕、移動時間が２０．０５〔ｍｉｎ〕、また、２回目の移動距離が１００２．４〔ｃｍ〕、移動時間が２０．０５〔ｍｉｎ〕であるとすると、最新の移動距離情報、及び、移動時間情報は今回までの移動距離、移動時間の平均を算出した移動距離９９４．３〔ｃｍ〕、移動時間１９．８９〔ｍｉｎ〕となる。移動距離、移動時間は過去の移動距離及び移動時間の平均を取ることにより、移動毎のばらつきを吸収した値が得られ、移動経路の選択を正確に行える。
【００９５】
なお、図２１では、過去から現在間での全ての情報の平均としたが、回数が多くなると、過去のデータを多数記憶していなければならず、過去のデータに支配され易くなり、現在の状況が反映されにくくなるなるので、現在から過去数回分のデータだけを保持し、その平均を最新のデータとする方法としてもよい。
【００９６】
図２２に本発明の一実施例の最新移動距離及び移動時間の算出方法の変形例を説明するための図を示す。
図２２では、現在ｎ回目の結果として、移動距離９９５．３〔ｃｍ〕、移動時間１９．９１〔ｍｉｎ〕が得られたとすると、現在のデータ移動距離９９５．３〔ｃｍ〕、移動時間１９．９１〔ｍｉｎ〕と過去２回分のデータ、すなわち、ｎ−２回目の結果、移動距離９８５．３〔ｃｍ〕、移動時間１４．４７〔ｍｉｎ〕、及び、ｎ−１回目の結果、移動距離１００２．４〔ｃｍ〕、移動時間２０．０５〔ｍｉｎ〕の平均が取られ、移動距離９９４．３〔ｃｍ〕、移動時間１９．８９〔ｍｉｎ〕が最新の移動距離、移動時間として設定される。
【００９７】
なお、過去の１回目からｎ−３回目までの結果は移動の度に削除される。
以上により、移動結果の保持データが過去２回分までよく、メモリ容量を削減できる。また、最近の情報だけが反映されるため、現在の状況に応じた値を得ることができる。
【００９８】
また、図２３に本発明の一実施例の存在確率の変更動作説明図を示す。
また、移動情報として存在確率が設定されていれば、通過の可・不可により存在確率が更新される。
存在確率Ｃｅは、常時通過可能であれば、「１」となり、通過できなければ「０」になり、一般に
【００９９】
【数２】

【０１００】
なお、Ｐは経験値で、初期状態では「０」で、このときの存在確率Ｃｅは「０．５」となり、一回通過できる度にインクリメントされ、一回通過できない度にデクリメントされ、存在確率Ｃｅが制御される。
式（２）の存在確率Ｃｅは、Ｐに対して図２３に示すような波形となる。例えば、１回通過できると、Ｐがインクリメントされ、「０．７３１０５９」、２回通過できると、さらに、Ｐがインクリメントされ「０．８８０７９７」、３回通過できると、さらに、Ｐがインクリメントされ、「０．９５２５７４」というように「１」に近似される。
【０１０１】
また、１回通過できない場合には、Ｐはデンクリメントされ、「０．２６８９４１」、２回通過できないと、さらに、Ｐがデンクリメントされ「０．１１９２０３」、３回通過できないと、さらに、Ｐがデンクリメントされ、「０．０４７４２６」というように「０」に近似される。
【０１０２】
以上により、移動情報が移動時に測定された、移動距離、及び、移動情報、移動の可否に応じて変更され、実際の移動に即した値にできる。このため、移動経路作成時に効率的な移動経路を正確に設定できるようになる。
図２０のステップＳ６−３で、以上のようにして移動情報の更新が終了すると、次に、ツアー群の配列を検索が効率よく行われるように並び換えられる。この処理が特許請求の範囲の請求項１６の実施例に相当する。
【０１０３】
図２４に本発明の一実施例のツアー群並べ替え処理の動作説明図を示す。図２４（Ａ）は移動前のツアー群、図２４（Ｂ）は移動後のツアー群を示す。
まず、今回移動で使用したツアーが図１０、図１４に示すようなツアー群に存在するか否かを検索する（ステップＳ６−４）。ステップＳ６−４で、今回移動で使用したツアーがツアー群に存在していれば、ツアー群中の各ツアーに設定された使用回数を「１」、インクリメントし、図２４に示すようにしよう回数順にツアーＡ〜Ｉの順に並べ替えられる（ステップＳ６−５、ステップＳ６−７）
また、ステップＳ６−４で、新規に作成されたツアーを用いた場合には、新規に作成されたツアーの使用回数を「１」として、ツアー群の配列を回数順に並べ替える（ステップＳ６−６、Ｓ６−７）。
【０１０４】
例えば、図２４（Ａ）に示すようにツアーＡ〜Ｉに検索順に配列されていたツアー群が使用したツアー図２４（Ａ）のツアーＢ（識別番号ＮＯ．３→ＮＯ．６）、ツアーＤ（識別番号ＮＯ．６→ＮＯ．１１）、ツアーＧ（識別番号ＮＯ．１→ＮＯ．２→ＮＯ．３）の使用回数に「１」がインクリメントされ、使用回数が大きい順に配列されることにより、図２４（Ｂ）に示すようにツアーＡに識別番号ＮＯ．３→ＮＯ．６、ツアーＤに識別番号ＮＯ．６→ＮＯ．１１、ツアーＥに識別番号ＮＯ．１→ＮＯ．２→ＮＯ．３の配列が位置し、検索順位が上昇される。
【０１０５】
このように、使用回数に応じて検索順位が上昇されることにより、移動経路作成時の検索時に必要とするツアーが早く検索できるため、移動経路作成を効率的に行える。
なお、ツアー群には、移動を実行した移動経路をそのまま、記録してもよい。
【０１０６】
図２５に本発明の一実施例のツアー群並べ替え処理の変形例の動作説明図を示す。図２５（Ａ）は移動前のツアー群、図２５（Ｂ）は移動後のツアー群を示す。
例えば、図２５（Ａ）に示すツアーＡ、Ｃ、Ｇ、Ｉを用いて識別番号ＮＯ．１→ＮＯ．２→ＮＯ．３→ＮＯ．６→ＮＯ．１１→ＮＯ．１２→ＮＯ．１３を通過する移動経路が作成され、実行されたとすると、図２５（Ｂ）に示すように実行された移動経路、識別番号ＮＯ．１→ＮＯ．２→ＮＯ．３→ＮＯ．６→ＮＯ．１１→ＮＯ．１２→ＮＯ．１３を検索順が最上位となるツアーＡに挿入し、以下、ツアーＢ〜Ｊに図２５（Ａ）のツアーを配列する。以上の処理が特許請求の範囲の請求項１１の処理に相当する。
【０１０７】
以上により、実行されたツアーが上位に位置することになるので、次回の移動時に、同一の移動開始位置、移動終了位置を設定されたときに、１回で移動経路が検索され、移動経路の作成を迅速に実行できる。
なお、中央演算処理装置７は、常時駆動用バッテリＢＡＴの容量を検出しており、バッテリＢＡＴの容量が所定のレベル以下に減少すると、図１２の経路計画作成処理のステップＳ４−１０で選択する移動経路を最短の移動経路を効率的な移動経路に変更する。
【０１０８】
以上により、移動速度や移動距離を少なくしなければならないときに、バッテリの状態を検出し、バッテリの容量に応じて移動経路を変更しすることができ、移動経路の途中で停止したり等の不具合を防止できる。
以上のように、本実施例によれば、環境が変更しても、地図を容易に変更できる。また、環境の細かい情報を意識せずに行動計画を建てることができるので、従来技術に比べてより現実的な時間で経路計画を行なうことができる。また、最初に正確な地図用意できなくても、実世界とのインタラクションにより実世界の情報を容易に取り込んで修正することが可能となる。
【０１０９】
【発明の効果】
上述の如く、本発明の 請求項１によれば、環境検出手段により周囲の環境を検出し、ランドマーク及びリンクからなる移動経路地図のランドマークを認識し、リンクの方向に順次移動することにより、ランドマーク及びリンクからなる移動経路地図上のランドマーク間の移動が可能となる。このとき、移動経路地図はランドマーク及びリンクだけの簡単な情報であるので、作成、変更が容易に行え、如何なる場所にも容易に適用でき、また、リンクの分岐するランドマークから分岐したリンク毎にランドマークの配列を生成し、データベースとして保持し、データベースに保持されたランドマークの配列を先頭のランドマークと後尾のランドマークとが一致するように選択し、順次接続することにより、ランドマークがリンクで接続された一連の移動経路を作成でき、このとき、分岐がないランドマークは、一通りの配列としてデータベースに保持されるため、データベースに保持するデータ量を削減でき、また、データベースに保持されたランドマークの配列を組み合わせることにより一連の移動経路を作成するので、移動経路の検索を容易に行える等の特長を有する。
【０１１０】
請求項２によれば、リンクが移動距離、移動時間、存在確率等のそのリンクの通過の難易度を示す移動情報をもつことにより、移動経路生成手段により生成された移動経路のうち、移動情報で、例えば、移動距離の総和が短い移動経路、移動時間の総和が短い移動経路、存在確率の高い移動経路を選択することにより、効率のよい移動経路を選択できる等の特長を有する。
【０１１１】
請求項３によれば、信号や踏切などの時間により通行可能、不可能となるもの経路を通過する時に、信号や踏切が通行可能となる時間には、信号や踏切を通過する経路を通り、信号や踏切が通行不可能となる時間には、信号や踏切を回避した経路を検索し、通る等の移動の効率が良い経路を検索できる。
【０１１２】
請求項４によれば、ランドマークに接続される前記リンクを順次追うことにより、ランドマークは通過地点であり、リンクは直接移動可能な経路であるので、ランドマークとリンクを接続した経路を検索することにより、移動すべき経路を容易に生成できる等の特長を有する。
【０１１４】
請求項５によれば、ランドマークに識別番号を付与し、付与された識別番号の配列としてデータベースが構成されるので、データベースを構成するデータ量を削減でき、また、識別番号の一致、不一致を検出するだけで、一連の移動経路を構成可能な配列を選択でき、移動経路の作成を容易に行える等の特長を有する。
【０１１５】
請求項６によれば、実際に移動した際の環境と地図情報とに差異があるときには、実環境に適合するように地図情報が変更される。例えば、地図情報作成時には存在しなかった障害物が地図情報で移動経路となっている部分に存在する場合などには、移動できないので、地図情報中のリンクを切断する等の処理を行い、次回の移動時にはその移動経路を通過する移動経路は選択されなくなり、実環境に適合した移動経路を作成できる等の特長を有する。
【０１１６】
また、地図情報作成時には存在した障害物が存在しなくなった場合等には、その部分に新たにランドマーク及びリンクを付加し、次回移動時には移動経路の一部として選択できるようにし、最適な移動経路を作成できるようできる等の特長を有する。
【０１１７】
請求項７によれば、環境検出手段で検出された移動経路の実際の環境に応じてリンクに付与された移動情報を変更することにより、地図情報を実環境により近いものにでき、次回の移動経路の作成時にはより実環境に即した移動経路が作成できる等の特長を有する。
【０１１８】
請求項８によれば、リンクに付与された移動情報を変更するときに、環境検出手段で検出された環境による移動情報の変更に過去にリンクに付与されていた移動情報を反映させることにより、環境検出手段で検出された環境のばらつきが吸収され、移動情報を実環境に近似させることができる等の特長を有する。
【０１１９】
請求項９によれば、移動経路として使われる頻度に応じてデータベースの検索順序を頻度の高い順に並べ替えることにより、移動経路を作成する際にデータベースの検索時間を短縮できる等の特長を有する。
請求項１０によれば、実際の移動した移動経路をデータベースに保存することにより、実際の移動した移動経路は新たに作成する必要がなく、また、一度作成した移動経路を元に新規の移動経路を生成することができるため、移動経路の作成時間を短縮できる等の特長を有する。
【図面の簡単な説明】
【図１】本発明の一実施例のブロック構成図である。
【図２】本発明の一実施例の地図情報の概略構成図である。
【図３】本発明の一実施例の地図作成処理の動作フローチャートである。
【図４】本発明の一実施例の地図情報生成処理の動作説明図である。
【図５】本発明の一実施例の地図情報生成処理のランドマーク設定時の動作説明図である。
【図６】本発明の一実施例のデータベース化処理の動作フローチャートである。
【図７】本発明の一実施例のデータベース化処理で識別番号を付与した状態を示す図である。
【図８】本発明の一実施例のデータベース化処理で付与される移動情報の構成図である。
【図９】本発明の一実施例のデータベース化処理で得られる移動情報の変形例の説明図である。
【図１０】本発明の一実施例のデータベース化処理で生成されるツアー群の構成図である。
【図１１】本発明の一実施例の移動処理の動作フローチャートである。
【図１２】本発明の一実施例の経路計画作成処理の動作フローチャートである。
【図１３】本発明の一実施例のツアー生成方法の動作説明図である。
【図１４】本発明の一実施例のツアー生成方法の動作説明図である。
【図１５】本発明の一実施例の地図情報更新処理の動作フローチャートである。
【図１６】本発明の一実施例のランドマーク及びリンク追加時の動作説明図である。
【図１７】本発明の一実施例のランドマーク及びリンク追加時の動作説明図である。
【図１８】本発明の一実施例のランドマーク及びリンク削除時の動作説明図である。
【図１９】本発明の一実施例のランドマーク及びリンク削除時の動作説明図である。
【図２０】本発明の一実施例の情報更新処理の動作フローチャートである。
【図２１】本発明の一実施例の最新移動距離及び移動時間の算出方法を説明するための図である。
【図２２】本発明の一実施例の最新移動距離及び移動時間の算出方法の変形例を説明するための図である。
【図２３】本発明の一実施例の存在確率の変更動作を説明するための図である。
【図２４】本発明の一実施例のツアー群並べ替え処理の動作説明図である。
【図２５】本発明の一実施例のツアー群並べ替え処理の変形例の動作説明図である。
【符号の説明】
１ 自律移動ロボット装置
２ 地図情報
３ 地図用メモリ
４ 環境センサ
５ 入出力装置
５ａ 入力装置
５ｂ ディスプレイ
６ インタフェース回路
７ 中央演算処理装置
８ プログラム用メモリ
９ 駆動回路
１０ 駆動機構[0001]
[Technical field to which the invention belongs]
The present invention relates to a travel route map, a travel route map generation method, and an autonomous mobile device, and more particularly to a travel route map, a travel route map generation method, and an autonomous device for setting a travel route of the mobile device. It relates to a mobile device.
[0002]
In recent years, robotic devices have become popular in the workplace and home. Most current robotic devices operate only in accordance with preset operations, and the setting must be changed when the operating environment changes. In the future, technology for adapting to constantly changing environments, such as the real world, will be indispensable for the spread of robots in the workplace and home.
[0003]
When the environment is not unknown at all, the robot can behave more quickly if there is no map because it is planned to have a map. Also, in the real world, you can only prepare an inaccurate map at first, and you will often want to correct it when you get to the place. Furthermore, in the real world, it is often difficult to know the distance or to know the distance accurately. Therefore, in order to move Ropot in the real world, it is considered indispensable to create a map corresponding to the changing environment.
[0004]
[Prior art]
Conventionally, a route design method in an autonomous mobile robot apparatus can be roughly divided into two flows: an artificial potential method and a graph search method.
In the artificial potential method, a potential is set for a robotic device so that the target is a suction force and the obstacle is a repulsive force, a potential space is created, and a path is created according to the shape of the created potential space. It is a technique to do.
[0005]
The graph search method is a method in which a space in which no obstacle exists is divided into a large number of cells, their connection status is expressed as a graph, and an optimum route is searched.
These two approaches have their merits and demerits, and in practice it is a natural way to use a combination of both. However, these methods are effective under a static and known environment, but become very complicated when they are dynamic, and only use locally when used in unknown environments. ing.
[0006]
Conventionally, a map is an absolute map created by a designer based on a behavior space that is predicted in advance. When a robot uses it, it is stored inside and moved according to the stored map. Here, an absolute map is a map that has a coordinate system such as latitude and longitude that is accurately described, such as a commercially available topographic map. In the case of an indoor map, such as a wall, door, or desk. The position is accurately described.
[0007]
For this reason, once a map is created, it takes a lot of effort to change the map, such as measuring the exact coordinates of the changed part. Therefore, in general, the map given to the robot at the time of design is not changed, and an action plan is always created based on the same map.
[0008]
In addition, although behavioral algorithms for creating maps in unknown spaces have been studied, there has been only research on behavioral spaces that have only a very simple shape, which is far from being in practical use.
[0009]
[Problems to be solved by the invention]
However, some of the information about the real-world environment that we are trying to use includes some that are known and some that are unknown. It is a very confused world where there are things that don't move and some things that move. In such an environment, it is natural to decide on the basis of characteristic things that do not change much and make an action plan based on them, and there is an absolute map as a map created in this way. In the conventional autonomous mobile device, the movement is performed based on the absolute map as described above.
[0010]
However, robots acting in the real world must take into account sensors that recognize the environment and errors in movement. Absolute maps originally do not take error into account, so they have the disadvantage of being difficult to use to correct and plan for measurement errors. However, there has been little research on action planning (path planning) corresponding to the real world in which known and unknown, static and dynamic objects are mixed.
[0011]
Also, the real world has so much information that I don't know what to use. Furthermore, it is difficult to accurately measure information such as distance and direction. When the environment is not known at all, the robot will be able to act more quickly than if there is no map, because it is planned to have a map. However, there is a problem that it is difficult to change the map with a normal map (absolute map) because it involves a lot of labor.
[0012]
The present invention has been made in view of the above points, and is a travel route map that can solve the above-described problems, can express uncertain spatial knowledge, and can flexibly cope with changes in the environment. An object of the present invention is to provide a moving route map generation method and an autonomous mobile device.
[0013]
[Means for Solving the Invention]
The present invention solves the above problems by expressing a map using landmarks. For example, people use maps that have inaccurate distances and directions, such as meal maps and subway maps, which are published in magazines, but have a place where they are featured. By recognizing the difference between the real environment and the map in my head, I can successfully reach the destination.
[0014]
This is because buildings and intersections are used as landmarks and the destinations are traced one after another. By using a certain thing as a landmark, it is possible to handle even if the information between them is slightly different. A building or scene existing in a real environment as a landmark when moving is called a landmark.
[0015]
A landmark map is a characteristic place of the environment, which is a map in which places that are convenient for realizing actions are used as landmarks and links between the landmarks are connected by links. In an actual environment, landmark information, such as landmarks, does not change significantly. Rather, the first time a map is created, it is likely to be inaccurate. Information such as distance. Therefore, in the landmark map, the landmarks are connected by links, and information such as the time and distance required for the links is given. In this way, even if the distance between certain landmarks is different from the initial value, it is only necessary to change the link information. On the other hand, in a normal map, if the distance is one place, you have to change everything.
[0016]
Further, since the map information described in this way consists only of landmarks and links connected thereto, changes such as addition / deletion of landmarks can be made relatively easily.
Therefore, the present invention uses a landmark map made up of landmarks and links, so that an autonomous mobile robot sequentially follows the landmarks along the links to achieve the intended action.
[0017]
  Claim 1 of the present invention is a moving device that moves under a changing environment, and a plurality of landmarks installed at movable positions in the moving environment, and directly among the plurality of landmarks Movement route map holding means for holding a movement route map composed of links connecting movable landmarks, and movement for generating a movement route based on the movement route map held in the movement map holding means A route generation unit; a movement control unit that controls movement based on the movement route generated by the movement route generation unit; an environment detection unit that detects the surrounding environment; and the detection unit that detects the surrounding environment. The environment state detected by the environment detection means when there is a difference between the environment state and the travel route map held in the travel route map holding means is the travel route location. It is reflected in by chromatic and travel route map update means for updating the travel route mapThe movement route map holding means holds a series of movement routes set as an array of a plurality of landmarks and links as divided movement routes divided for each branch of the link, and the movement route generation means The desired movement path is generated by sequentially connecting the divided movement paths.It is characterized by that.
[0018]
In the second aspect of the present invention, the movement route map holding unit holds movement information indicating a difficulty level when moving between the landmarks connected for each link,
The movement route generation means generates a plurality of movement routes and selects a route having a small sum of the movement information.
[0019]
According to a third aspect of the present invention, the movement route map holding unit sets the movement information according to an absolute time,
The moving device according to claim 1 or 2, wherein the moving route generation unit selects a moving route having a small sum of the absolute times from the generated moving routes.
[0020]
According to a fourth aspect of the present invention, the moving route map holding unit holds a series of moving routes as an array of the plurality of landmarks,
The movement path generation means generates a desired movement path by combining the landmark arrangement.
[0022]
  Claim5The moving route map holding unit holds the series of moving routes as an array of identification numbers assigned in advance to the plurality of landmarks, and the moving route generation unit generates the moving route with the array of identification numbers. The movement control means controls movement so as to sequentially follow the identification numbers arranged by the movement path generation means.
[0023]
  Claim6If the location where the travel route map update means can move on the travel route map held by the travel route map holding means is unmovable as a result of detection by the environment detection means In the above, the landmark and the link connected to the place where the movement route map held by the movement route map holding unit cannot move are deleted, and the movement route map holding unit holds the If the location that is impossible to move on the movement route map is movable as a result of detection by the environment detection means, the movement route map held in the movement route map holding means can be moved.Noh andThe landmarks and the links are added to all the places that have become.
[0024]
  Claim7Has movement information update means for detecting the movement information by the environment detection means during movement, and updating the movement information according to the movement result after movement.
  Claim8Is characterized in that past data is reflected when the movement information updating means updates the movement information.
[0025]
  Claim9Detects the frequency with which the series of travel routes is used as a travel route, and determines the series of travel routes in the database provided in the travel route map holding unit according to the frequency detected by the frequency detection means. It has a database changing means for sequentially rearranging in the order of access.
[0026]
  Claim10Has database control means for holding the movement path actually moved in the database as the series of movement paths.
  Next, the operation of each claim will be described. According to claim 1, the environment detection means detects the surrounding environment, recognizes the landmark of the movement route map composed of the landmark and the link, and sequentially moves in the direction of the link, thereby comprising the landmark and the link. It is possible to move between landmarks on the travel route map. At this time, since the travel route map is simple information of only landmarks and links, it can be easily created and changed, and can be easily applied to any location.Also, an array of landmarks is generated for each link branched from the landmark where the link branches, and is stored as a database, and the landmarks stored in the database match the first landmark and the last landmark. A series of movement paths in which landmarks are connected by links can be created by selecting and sequentially connecting them. At this time, landmarks without branches are held in the database as a single array, so the amount of data held in the database can be reduced. In addition, since a series of travel routes is created by combining the arrangement of landmarks held in the database, the travel route can be easily searched.
[0027]
According to claim 2, the movement information of the movement route generated by the movement route generation means by having the movement information indicating the difficulty of passage of the link such as the movement distance, the movement time, and the existence probability of the link. Thus, for example, an efficient travel route can be selected by selecting a travel route with a short total travel distance, a travel route with a short total travel time, and a travel route with a high existence probability.
[0028]
According to claim 3, when passing through a route that can pass or cannot pass by a time such as a signal or a railroad crossing, the signal or a railroad crossing passes through a route passing through the signal or a railroad crossing, During times when traffic lights and railroad crossings are not allowed, a route that avoids traffic lights and railroad crossings can be searched, and a route with good movement efficiency such as passing through can be searched.
[0029]
According to claim 4, by sequentially following the links connected to the landmark, the landmark is a passing point, and the link is a directly movable route, so the route connecting the landmark and the link is searched. By doing so, a route to be moved can be easily generated.
[0031]
  Claim5According to the above, the identification number is assigned to the landmark, and the database is configured as an array of the assigned identification numbers, so the amount of data constituting the database can be reduced, and only the identification number match or mismatch is detected. Thus, it is possible to select an array that can constitute a series of movement paths, and to easily create a movement path.
[0032]
  Claim6According to the above, when there is a difference between the environment when actually moved and the map information, the map information is changed to match the actual environment. For example, if an obstacle that did not exist at the time of map information creation exists in the part that is the travel route in the map information, it can not move, so processing such as cutting the link in the map information is performed next time During the movement, a movement route that passes through the movement route is not selected, and a movement route suitable for the actual environment can be created.
[0033]
  Also, when there are no obstacles that existed at the time of map information creation, a new landmark and link are added to that part so that it can be selected as a part of the movement route at the next movement, and the optimum movement You can create a route.
  Claim7The map information can be made closer to the actual environment by changing the movement information given to the link according to the actual environment of the movement route detected by the environment detection means, and the next movement route is created. Sometimes it is possible to create a movement route that is more suited to the actual environment.
[0034]
  Claim8When the movement information given to the link is changed, the environment detection means is reflected by reflecting the movement information previously given to the link in the change of the movement information by the environment detected by the environment detection means. Variations in the environment detected inThe injuryIt is absorbed and the movement information can be approximated to the real environment.
[0035]
  Claim9According to the above, the database search time can be reduced when creating a travel route by rearranging the database search order in the order of frequency according to the frequency used as the travel route.
  Claim10According to the above, it is not necessary to newly create the actual travel route by storing the actual travel route in the database, and a new travel route is generated based on the travel route once created. Therefore, the travel route creation time can be shortened.
[0036]
DETAILED DESCRIPTION OF THE INVENTION
The present invention relates to map information in an autonomous mobile device, and is roughly classified into three. The first is to express the environment as a map using landmarks, the second is to realize movement using the map, and the third is to change the landmark map when the environment changes In the point.
[0037]
First, an autonomous mobile robot apparatus equipped with map information will be described.
The point of expressing the first environment with a map using landmarks will be described.
FIG. 1 is a block diagram of an autonomous mobile robot according to an embodiment of the present invention.
The autonomous mobile robot device 1 of this embodiment corresponds to the autonomous mobile device of the claims, and includes a map memory 3 for storing map information 2 including landmarks and links, and environment detection for detecting the surrounding environment. Sensor 4, input / output device 5 for performing various instructions for creation and movement of map information 2, environment detection sensor 4, input / output device 5, interface circuit 6 for input / output control of drive signals, map information 2, and a drive control signal for moving according to the movement plan based on the program from the map information 2 stored in the map memory 3 and the surrounding environment information detected by the environment detection sensor 4 The central processing unit 7 is connected to a program memory 8 in which a program to be processed by the central processing unit 7 is stored, and the interface circuit 6. 7 is supplied via the interface circuit 6 and the drive circuit 9 generates a drive signal for movement based on the drive control signal supplied from the interface circuit 6. The drive signal generated in the above is supplied, the motor is driven in accordance with the supplied drive signal, and the entire apparatus is freely moved on a plane. The battery BAT supplies the drive power to the drive mechanism 10. Is done.
[0038]
The map memory 3 stores map information 2 composed of landmarks and links, which is a feature of the present invention.
Here, the map information 2 stored in the map memory 3 will be described.
FIG. 2 is a schematic configuration diagram of map information according to an embodiment of the present invention.
[0039]
The map information 2 of the present embodiment corresponds to the travel route map in the claims, and is used for instructing the travel route of the autonomous mobile robot device 1, and is a landmark 11 that is a fixed position on the travel route. -1 to 11-n and landmarks 11-1 to 11-n, the landmark 11-1 and the landmark 11-5, the landmark 11-2 and the landmark 11- 6, the links 11-1 to 12-m connecting the landmark 11-5 and the landmark 11-6.
[0040]
The landmarks 11-1 to 11-n correspond to the landmarks in the claims. For example, if the movement route is an office, a corner or a branch point of a corridor, a center of a room, an entrance of a room door, It is set at a position where an autonomous mobile robot device such as an exit can pass. The positions of the landmarks 11-1 to 11-n are set by coordinates (X, Y).
[0041]
The links 12-1 to 12-m correspond to the links in the claims, have a straight line shape, and the autonomous mobile robot device 1 can directly move linearly among the landmarks 11-1 to 11-n. .. Are connected between the landmark 11-1 and the landmark 11-5, the landmark 11-2 and the landmark 11-6, and the landmark 11-5 and the landmark 11-6. The links 12-1 to 12-m are assigned and set with distance (travel) and time movement information for moving between landmarks to which the autonomous mobile robot device 1 is connected. This movement information corresponds to movement information indicating the difficulty level when moving within the scope of claims.
[0042]
Here, the structure of the movement information according to claims 5 to 8 and the method for generating the map information 2 corresponding to claims 9 to 11 will be described. FIG. 3 is an operation flowchart of map information generation processing according to an embodiment of the present invention, FIG. 4 is an operation explanatory diagram at the time of landmark setting of map information generation processing according to the present invention, and FIG. 5 is map generation according to an embodiment of the present invention. FIG. 6 shows an operation explanatory diagram of the database creation processing of the map generation processing according to the embodiment of the present invention. 4A is a layout map of the environment, and FIG. 4B is a diagram showing how landmarks are set.
[0043]
When generating the map information 2, the user generates the map information 2 in the map memory 3 using the input / output device 5 of the autonomous mobile robot device 1.
When the input device 5a of the input / output device 5 is operated by the user and the map information generation is instructed by the user, the autonomous mobile robot device 1 applies the map information generation processing program stored in the program memory 8 and accordingly the central processing processing. The device 7 is controlled.
[0044]
When the map information generation processing program is started, the central processing unit 7 instructs the display 5b of the input / output device 5 to input a simple layout map indicating the environment in which the autonomous mobile robot device 1 is moved (step S1- 1). In response to a layout map input request from the autonomous mobile robot device 1, the user inputs a layout map as shown in FIG. 4A, for example. At this time, the layout map input by the user is a map that helps the user to set the landmark, and the position where the autonomous mobile robot device 1 can pass as shown in FIG. It only needs to be able to recognize the outline, and it can be very simple. If there is no layout map, the landmark position may be input as coordinates in a moving environment.
[0045]
When the layout diagram is input in step S1-1, the central processing unit 7 displays the layout diagram on the display 5b and instructs the user to input a landmark (step S1-2).
When the user operates the input device 5a in accordance with an instruction from the autonomous mobile robot device 1, for example, as shown in FIG. 4B, a landmark R indicating the center of the room on the layout map, a door room The landmark Do indicating the outside of the door, the landmark Di indicating the inside of the door room, the landmark C indicating the turning corner of the moving path, and the landmark J indicating the branch point of the moving path are arranged at positions corresponding to the respective parts of the layout diagram. .
[0046]
When the landmark setting is completed, the user operates the input device 5a to give an instruction to the effect that the landmark setting has been completed. When the central processing unit 7 is supplied with an instruction to the effect that the landmark setting has been completed from the input device 5a, the central processing unit 7 then displays a link setting instruction on the display 5b (steps S1-3 and S1-4). .
[0047]
When the user operates the input device 5a according to an instruction from the autonomous mobile robot device 1, for example, links 12-1 to 12-15 are set as shown in FIG. At this time, the links 12-1 to 12-15 are represented by straight lines, and the user can move, for example, the landmark 11-1 and the landmark 11-2, and the landmark 11-2 and the landmark 11-3 that can move along the straight line. .., Etc. need to be connected by links 12-1, 12-2, 12-3,.
[0048]
Next, when the link setting is completed, the user operates the input device 5a to instruct that the landmark setting is completed. When the central processing unit 7 receives an instruction from the input device 5a to the effect that the link setting has been completed, the central processing unit 7 then creates a database corresponding to the supplied landmark and link, and ends the map information generation process. (Steps S1-6, S1-7).
[0049]
Next, the database creation process of the map information generation process in step S1-6 will be described.
FIG. 6 is an operation flowchart of database creation processing according to an embodiment of the present invention, FIG. 7 is a diagram illustrating a state in which an identification number is assigned in the database creation processing according to one embodiment of the present invention, and FIG. FIG. 9 is a diagram for explaining the existence probability assigned in the database creation process of the embodiment of the present invention, and FIG. 10 is a diagram of the embodiment of the present invention. The block diagram of the tour group produced | generated by database formation process is shown.
[0050]
In the database processing, the central processing unit 7 first assigns identification numbers (IDs) to the set landmarks 11-1 to 11-15 in the order of setting the landmarks in step S1-2 as shown in FIG. ) NO. 1-NO. 15 is added (step S2-1). This identification number corresponds to the identification information in the claims.
[0051]
Next, the central processing unit 7 sets movement information to the links 12-1 to 12-16 connecting the landmarks 11-1 to 11-15 with the IDs (step S2-2). The movement information corresponds to the movement information in the claims. As shown in FIG. 8A, the links 12-1 to 12-16 are assigned identification numbers NO.1 to 11-11 of the landmarks 11-1 to 11-15. 1-NO. 15, for example, the link 12-1 connecting the landmark 11-1 and the landmark 11-2 is identified by the identification number NO. 1-NO. 2, the link 12-2 connecting the landmark 11-2 and the landmark 11-3 has an identification number NO. 2-NO. 3 is identified, and the distance (movement distance) a, time (movement time) b, existence probability c, and the like are set for each of the links 12-1 to 12-16.
[0052]
The movement distance a of the movement information corresponds to the movement distance in the claims, and is set by obtaining the straight line distance L from the position (coordinates) of the landmark. For example, if the coordinate of the landmark 11-1 is (0, 0) and the coordinate of the landmark 11-2 is (100, 0), the movement distance L is “100 [cm]”. Further, the time b of the movement information is obtained by L / V from the linear distance L between predetermined landmarks and the movement speed V of the autonomous mobile robot device 1. For example, if the movement distance L between the landmark 11-1 and the landmark 11-2 is "100 [cm]" and the movement speed V of the autonomous mobile robot device 1 is "0.5 [cm / sec]", the movement The time b is “200 [sec]”.
[0053]
The existence probability c of the movement information indicates the existence probability of the links 12-1 to 12-16, and is “1” if it can pass, and “0” if it cannot pass. Thus, “0.5” is set when the passage is half and the passage is impossible.
[0054]
The existence probability c is determined when there is an obstacle that can pass through or cannot pass through time, such as a signal, a railroad crossing, and a door that opens and closes depending on the time, on the links 12-1 to 12-16. It becomes effective. That is, if the existence probability c is “0”, it is possible to pass, and if the existence probability c is “1”, it can be judged that the passage is impossible.
[0055]
In FIG. 8A, the case where the movement distance a, the movement time b, and the existence probability c are fixed has been described. However, as shown in FIG. 8B, the movement time b and the existence probability c are used as functions of time. Conceivable configuration.
FIG. 9 shows an explanatory diagram of a modified example of the movement information obtained by the database processing according to the embodiment of the present invention.
[0056]
For example, the existence probability c is set in advance between “0” and “1” as shown in FIG. 9A according to the function f (T) of the absolute time T set in the central processing unit 7. Change the value at The existence probability c is set to “0.5” in the initial state, and is updated according to the movement result as will be described later. The updated value is approximated to the actual value.
[0057]
The existence probability c at this time is effective, for example, when the door opening / closing frequency changes according to the time zone when the door is opened / closed. For example, when the moving route is generated, the moving time can be shortened by referring to the existence probability c and selecting a route with a high probability that the door is open.
[0058]
If there is an obstacle such as a signal that can be passed or not passed at a constant cycle, the probability is “1” or “0” at a constant cycle as shown in FIG. 9B. And the period of the probabilities “0” and “1” may be changed by changing the parameters of the function according to the actual waiting time at the time of passing.
[0059]
Note that the travel time b also changes depending on the time depending on the link signal, railroad crossing, etc., and therefore can be expressed as a function of the absolute time T as with the existence probability c as shown in FIG. 9C.
When movement information as shown in FIG. 8 is obtained in step S2-2, a tour group is then generated. The tour group is obtained by dividing each map information branch point.
[0060]
As shown in FIG. 10, the tour group includes landmark identification numbers NO. 1-NO. It is set as 15 arrays.
For example, when the landmark is arranged as shown in FIG. The landmark No. 1 has the identification number NO. No. 2 landmark, and the identification number NO. The landmark No. 2 has the identification number NO. No. 3 landmark only, and the identification number NO. The landmark No. 3 is the identification number NO. 4 and identification number NO. Since the branch is branched to the landmark No. 6, the uniquely determined route has the identification number NO. 1 → ID No. 2 → Identification number NO. Only 3 sequences. This array is registered as tour A.
[0061]
Also, the identification number NO. The landmark No. 3 is the identification number NO. 4 landmark and identification number NO. Since the route branches to the landmark No. 5, the identification number NO. Two new tours B and C are formed from the three landmarks. Tour B is the branch of the landmark with the next smallest identification number. The sequence uniquely determined by the branch is the identification number NO. 3 → ID No. 4 → ID No. 5 Tour C has an identification number of NO. 3 is a branch having a larger identification number, and the sequence uniquely determined by the branch is the identification number NO. 3 → ID No. 6
[0062]
As described above, by dividing the tour for each branch, tours A to I are generated as shown in FIG. Tours A to I as shown in FIG. 10 are stored in the map memory 3 as a database. The central processing unit 7 creates a series of travel routes by combining the tours A to I when creating a travel route at the time of a movement instruction. At this time, the central processing unit 7 searches for tours in the order of tours A → B → C → D → E → F → G → H → I and determines a combination.
[0063]
Thus, the generation of the database stored in the map memory 3 is completed.
Next, the case where the autonomous mobile robot apparatus 1 is moved according to the database stored in the map memory 3 will be described.
FIG. 11 shows an operation flowchart of the movement process according to the embodiment of the present invention.
[0064]
When the user operates the input device 5a to input a movement command, the central processing unit 7 detects this, and then performs a route plan creation process for setting the movement route of the autonomous mobile robot device 1 (step) S3-1, S3-2).
Here, the route plan creation process will be described.
[0065]
FIG. 12 shows an operation flowchart of a route plan creation process according to an embodiment of the present invention.
The route planning process corresponds to the embodiments of claims 4 and 5.
In the route planning process, the central processing unit 7 first instructs the user to input the landmark identification number IDs of the movement start position and the landmark identification number IDg of the movement end position (step S4-1). ). The central processing unit 7 receives an instruction to input the landmark identification number IDs of the movement start position and the landmark identification number IDg of the movement end position from the user (step S4-2). The tour group stored in the map memory 3 is searched, and the tour having the landmark identification number IDs of the movement start position at the head and the landmark identification number IDg of the movement end position at the end is searched (step S4). -3).
[0066]
As a result of the search in step S4-3, if there is a tour in which the landmark identification number IDs of the movement start position is the first and the landmark identification number IDg of the movement end position is the last, there is a landmark identification number of the movement start position. The identification number IDg of the landmark with the IDs at the head and the end position of movement is determined as the travel route, and the landmark ID IDs of the landmark at the movement start position is the head and the landmark at the movement end position is identified. If there is no tour whose number IDg is at the end, next, the tour whose landmark identification number IDs at the movement start position is the head or whose landmark identification number IDg is the tail at the movement end position is searched (step). S4-4, S4-5).
[0067]
As a result of the search in step S4-5, the landmark identification number IDs of the searched movement start position is the landmark identification number IDsg of the end of the first tour and the landmark identification number IDg of the searched movement end position. Are compared with the first IDgs to determine whether there is a match (step S4-6).
[0068]
In step S4-6, the landmark identification number IDs at the movement start position is compared with the landmark identification number IDsg at the end of the first tour and the first IDgs of the landmark identification number IDg at the movement end position. If there is no match, a tour connecting the landmark identification number IDs of the movement start position and the landmark identification number IDg of the movement end position cannot be obtained. As the mark identification number IDs, the landmark identification number IDsg at the end of the first tour is the first identification number IDs, and the first identification number IDgs of the landmark identification number IDg at the movement end position is the last identification number IDg. Returning to step S4-3, steps S4-3 to S4-6 are repeated, and the landmark identification number IDs0 of the movement start position and the movement end Determine the identification number IDg0 and a tour that can be connected Landmarks location (step S4-7).
[0069]
In step S4-6, the landmark identification number IDs at the movement start position is compared with the landmark identification number IDsg at the end of the first tour and the first IDgs of the landmark identification number IDg at the movement end position. As a result, if there is a match, the landmark identification number IDs at the movement start position can be linked with the tour with the landmark identification number IDg at the head and the landmark identification number IDg at the movement end position. Then, a series of tours is generated in which the landmark identification number IDs at the movement start position is at the head and the landmark identification number IDg at the movement end position is at the end (step S4-8). The tour search method in steps S4-3 to S4-8 is called a so-called graph search method.
[0070]
When a plurality of tours are generated in step S4-8, the optimum tour is selected using the movement information as shown in FIG. 8 (steps S4-9, S4-10).
For the selection of the optimum tour, for example, a tour with a small sum of movement distance a or movement time b is selected from the movement information set in the generated tour link.
[0071]
FIG. 13 and FIG. 14 are diagrams for explaining the operation of the tour generation method according to one embodiment of the present invention.
For example, the identification number NO. 1-NO. Assume that a map composed of 26 landmarks is generated, and a tour group as shown in FIG. 14 is generated from the map of FIG. Here, the identification number of the movement start position is set to NO. 1. The identification number of the movement end position is NO. If it is set to 20, the identification number NO. The tour in which 1 is located at the end is only the tour indicated by A in FIG. Since there are only L tours in which 20 is located at the end, tours A and L are selected first.
[0072]
However, the identification number at the end of Tour A is NO. 3 and the identification number at the beginning of tour L is NO. 18, tour A and tour L cannot be directly connected because they do not match. Next, the identification number NO. 3 searches for the first tour. Identification number NO. The tour with 3 at the end is B and Q, and the identification number NO. 3, the route from Tour A branches to Tour B and Tour Q.
[0073]
Similarly, if the search is continued, the identification number NO. 1 to the identification number NO. As a route to 20, tour A → Q → O → M → L (identification number NO.1 → NO.2 → NO.3 → NO.24 → NO.23 → NO.21 → NO.18 → NO.19 → NO.20) and tour A → B → D → F → H → L (identification number NO.1 → NO.2 → NO.3 → NO.4 → NO.5 → NO.8 → NO.11) → NO.12 → NO.15 → NO.18 → NO.19 → NO.20) Two routes are generated. If the sum of the travel distance a of the links of tour A → Q → O → M → L is smaller than the sum of the travel distance a of the links of tour A → B → D → F → H → L, then tour A → Q. → O → M → L is selected.
[0074]
The value to be compared with the selection is not limited to the total of the movement distance a, but may be, for example, the total of the movement time b. Furthermore, in order to improve the reliability, the movement distance a, the movement time b, Evaluation function J with existence probability C as a variable
[0075]
[Expression 1]

[0076]
May be set in advance, and may be selected according to the magnitude relationship of values obtained from the evaluation function.
Returning to FIG. 11, the description will be continued. When the movement route is generated by the above process in step S3-2, the central processing unit 7 starts the movement process and starts moving according to the movement route generated in step S3-2 (step S3-3). . At this time, the central processing unit 7 reads the identification number in order from the tour of the selected travel route, reads the landmark information corresponding to the identification number from the read identification number, and reads the landmark information and The movement is realized by following the landmarks one after another while comparing with the object detected and extracted by the environmental sensor 4 and determining whether or not it is the target landmark.
[0077]
The environmental sensor 4 is composed of, for example, an ultrasonic sensor or an infrared sensor, and captures the surrounding situation as an image. The central processing unit 7 analyzes the image captured by the environment sensor 4, recognizes an object (door, corner, room), etc. from the image, and determines the surrounding situation (step S3-4). The ambient condition detected by the environment center 4 is not limited to an image. In short, it is sufficient that a landmark can be identified, and a simple sonar sensor, a range finder contact sensor, or the like may be used. That is, the information obtained from the sensor may be abstracted and compared with a sensor image such as a landmark (prototype door) that is known in advance.
[0078]
At this time, if an environment change different from the map information on the travel route, for example, if an obstacle or the like is found on the route to be traveled, it is impossible to pass as it is, so map information update processing for changing the route is performed. Perform (step S3-5). Step S3-5 executes processing corresponding to claims 14, 15, and 17.
[0079]
The process moves to the final destination while repeating steps S3-3, S3-4, and S3-5 (step S3-6).
FIG. 15 shows an operational flowchart of the map information update process according to the embodiment of the present invention.
First, the central processing unit 7 analyzes the detected environmental change and determines whether or not the movement route is affected (steps S5-1, S5-2, and S5-3). For example, when the opening is detected by the environment sensor 4, the size of the opening is compared with the size of the device itself, and it is realized by determining whether or not the passage is possible. Further, when an obstacle is detected by the environment sensor 4, if the obstacle is on the movement route, the obstacle, the movement route, the interval, and the size of the own device are compared to detect whether the vehicle can pass on the movement route. It is realized by.
[0080]
When a new passable path is found in step S5-2, the central processing unit 7 adds a landmark and a link (steps S5-4 and S5-5).
FIGS. 16 and 17 are diagrams for explaining the operation of adding landmarks and links according to an embodiment of the present invention. 16A shows the arrangement of landmarks, FIG. 16B shows a map in which landmarks of map information are converted into identification numbers, FIG. 17A shows tour groups before addition, and FIG. 17B shows additions. The following tour group is shown.
[0081]
If a new entrance / exit 32 is recognized in the room 31 as shown in FIG. 16 (A), the central processing unit 7 first sets a new landmark Di in the inner part of the room 32 of the entrance / exit 32, Set a new landmark Do. Next, as shown in FIG. 16 (B), the preset identification numbers NO. 1-NO. 15 followed by identification number NO. 16, NO. 17 is set, and links 33 and 34 are added. Further, as shown in FIG. 8, the movement distance a and the movement time b are calculated from the coordinates and given to the links 33 and 34 from the coordinates, and the existence probability is set to the initial value 0.5.
[0082]
At this time, the identification number NO. 16 is an identification number NO. 11 and NO. 14 is located on the link connecting the ID number 11 and the NO. 16 and the identification number NO. 14 and NO. 16 may be divided into links 36 linking to 16.
[0083]
As described above, the identification number NO. 16, NO. When 17 landmarks and links are added, the data of the tour group shown in FIGS. 10 and 14 is changed. Identification number NO. 16, NO. Before the 17 landmarks and links are added, as shown in FIG. 7 → NO. 8 → NO. 9 was the identification number NO. 16, NO. As shown in FIG. 17B, the identification number NO. 7 → NO. 8 → NO. 9 → NO. 17 → NO. 16 In addition, the identification number NO. 11 → NO. 14 is an identification number NO. Of Tour H shown in FIG. 11 → NO. 16 and tour I identification number NO. 14 → NO. 16 is added.
[0084]
Further, when a route that cannot be passed is found in step S5-2, the central processing unit 7 deletes the landmark and the link (steps S5-6 and S5-7).
FIG. 18 and FIG. 19 are explanatory diagrams of landmarks and operations at the time of deletion according to an embodiment of the present invention. 18A shows the arrangement of landmarks, FIG. 18B shows a map in which landmarks of map information are converted into identification numbers, FIG. 19A shows a tour group before deletion, and FIG. 19B shows deletion. The following tour group is shown.
[0085]
As shown in FIG. 18 (A), when an obstacle is placed on the movement route and traffic is impossible, the identification number NO. 6 landmark and identification number NO. No traffic between the landmark No. 7 and the identification number NO. 6 landmark and identification number NO. The link connecting 7 landmarks is cut.
[0086]
Identification number NO. 6 landmark and identification number NO. When the link connecting the 7 landmarks is cut, the tour group information shown in FIGS. 10 and 14 is changed as in the case of addition. Identification number NO. 6, NO. Before the link between 7 is deleted, the tour D set as shown in FIG. 19A is deleted as shown in FIG. 19B.
[0087]
As described above, landmarks, links, and deletions relating to routes that have become passable are added, and the creation of a tour group is completed.
Returning to FIG. 15 and continuing the description, if landmarks, links are added and deleted as described above, it may not be possible to pass, so based on the newly created tour group, A route plan is created again (step S5-8). The route plan creation has already been described with reference to FIG.
[0088]
As described above, when an environmental change is detected by the environmental sensor 4 during movement, the landmark and the link are rewritten according to the detected environment, the tour group is changed, a new route is created, and the movement is performed. Continue.
According to the above processing, the change of the landmark map occurs when there is a place where the robot can make a new place and where it can no longer be done, and if necessary, add and delete landmarks, Further, the database tour group is reconfigured by creating and deleting links. Such a change is much easier than changing the map indicated by the absolute position.
[0089]
In addition, such a change is performed only when the autonomous mobile robot device 1 detects a change in the environment and only when the behavior of the autonomous mobile robot device 1 is affected. That is, small environmental changes unrelated to the movement of the autonomous mobile robot device 1 and slight sensor errors can be ignored, and the amount of calculation in the central processing unit 7 can be reduced.
[0090]
Here, returning to FIG. 11, the description will be continued.
At this time, the central processing unit 7 measures and stores the movement distance and the movement time, and stores the movement distance and the movement time in the map memory 3 according to the measured movement distance and movement time after the movement is completed. Update movement information stored in a database and give accurate information.
[0091]
When arriving at the final destination in step S3-6, the central processing unit 7 ends the robot movement process and performs an information change process for changing the movement information or the like according to the held information (step S3). -7). This step S3-7 corresponds to the embodiments of claims 12, 13, 14, 15, and 16.
[0092]
FIG. 20 shows an operation flowchart of the information update process according to the embodiment of the present invention.
The information update processing corresponds to the embodiments of claims 8 and 9 in the claims.
In the information update process, the central processing unit 7 acquires the movement distance and movement time between each landmark recorded during movement (step S6-1). Next, the central processing unit 7 acquires the movement distance, movement time, movement distance at the time of past movement, and the latest movement distance, movement time, existence probability acquired in step S6-1. Is calculated and updated (steps S6-2 and S6-3).
[0093]
FIG. 21 is a diagram for explaining a method of calculating the latest travel distance and travel time according to an embodiment of the present invention.
As a method of calculating the latest travel distance and travel time, there is a method of taking an average value with past data.
[0094]
For example, the identification number NO. 1 landmark and identification number NO. As a result of the first movement between two landmarks, it is assumed that the movement distance is 995.3 [cm] and the movement time is 19.91 [min]. At this time, the current movement is the third time. At this time, the first moving distance is 985.3 [cm], the moving time is 20.05 [min], the second moving distance is 1002.4 [cm], and the moving time is 20.05 [min]. Is the latest travel distance information and travel time information, the travel distance up to this time, the travel distance 994.3 [cm] calculated from the average travel time, and the travel time 19.89 [min]. As the movement distance and movement time, the average of the past movement distance and movement time is taken to obtain a value that absorbs the variation for each movement, and the movement route can be selected accurately.
[0095]
In FIG. 21, the average of all information from the past to the present is used. However, as the number of times increases, a large number of past data must be stored, and it becomes easy to be controlled by the past data. Since the situation is less likely to be reflected, only the past several times of data from the present may be retained, and the average may be used as the latest data.
[0096]
FIG. 22 is a diagram for explaining a modification of the latest moving distance and moving time calculation method according to the embodiment of the present invention.
In FIG. 22, assuming that the moving distance 995.3 [cm] and the moving time 19.91 [min] are obtained as a result of the current nth time, the current data moving distance 995.3 [cm] and the moving time 19. 91 [min] and data for the past two times, that is, the result of the n-2th time, the moving distance 985.3 [cm], the time of movement 14.47 [min], and the result of the n-1 time, the moving distance 1002 .4 [cm] and travel time 20.05 [min] are averaged, and travel distance 994.3 [cm] and travel time 19.89 [min] are set as the latest travel distance and travel time.
[0097]
The past results from the first time to the n-3th time are deleted each time they move.
As described above, the retention data of the movement result is sufficient up to the past two times, and the memory capacity can be reduced. Moreover, since only recent information is reflected, a value according to the current situation can be obtained.
[0098]
FIG. 23 is a diagram for explaining the operation of changing the existence probability according to the embodiment of the present invention.
If the existence probability is set as the movement information, the existence probability is updated depending on whether the passage is possible.
The existence probability Ce is “1” if it can always pass, and “0” if it cannot pass,
[0099]
[Expression 2]

[0100]
Note that P is an empirical value, which is “0” in the initial state, and the existence probability Ce at this time is “0.5”, which is incremented every time it can pass once, decremented every time it cannot pass, and the existence probability Ce is controlled.
The existence probability Ce in Expression (2) has a waveform as shown in FIG. For example, if it can pass once, P is incremented, “0.731059”, if it can pass twice, P is further incremented and “0.880797”, if it can pass three times, P is further incremented, It approximates to “1” such as “0.952574”.
[0101]
Also, if it cannot pass once, P will be decremented, “0.268941”, if it cannot pass twice, then P will be decremented, “0.119203”, if it cannot pass three times, then P Is decremented and approximated to “0” such as “0.047426”.
[0102]
As described above, the movement information is changed according to the movement distance, movement information, and availability of movement measured at the time of movement, and can be set to a value according to actual movement. For this reason, an efficient movement route can be accurately set at the time of movement route creation.
When the update of the movement information is completed as described above in step S6-3 in FIG. 20, the tour group arrangement is then rearranged so that the search can be performed efficiently. This process corresponds to the embodiment of claim 16 in the claims.
[0103]
FIG. 24 is a diagram for explaining the operation of the tour group rearrangement process according to the embodiment of the present invention. FIG. 24A shows a tour group before movement, and FIG. 24B shows a tour group after movement.
First, it is searched whether or not the tour used in the current movement exists in the tour group as shown in FIGS. 10 and 14 (step S6-4). In step S6-4, if the tour used for the current movement exists in the tour group, the use count set for each tour in the tour group is incremented by "1", and the number of attempts to be made as shown in FIG. The tours are rearranged in the order of tours A to I (step S6-5, step S6-7).
If a newly created tour is used in step S6-4, the number of times of use of the newly created tour is set to “1”, and the arrangement of tour groups is rearranged in order of number of times (step S6-6). , S6-7).
[0104]
For example, as shown in FIG. 24 (A), tour B (identification number NO.3 → NO.6) in tour FIG. 24 (A) used by tour groups arranged in the search order in tours A to I, tour D (Identification number NO.6 → NO.11), “1” is incremented to the number of times of use of tour G (identification number NO.1 → NO.2 → NO.3), and the number of times of use is arranged in descending order. , As shown in FIG. 3 → NO. 6. Tour D has an identification number NO. 6 → NO. 11. Tour E with identification number NO. 1 → NO. 2 → NO. 3 sequences are located and the search order is increased.
[0105]
In this way, by increasing the search order according to the number of times of use, it is possible to quickly search for a tour required at the time of searching at the time of creating a moving route, so that the moving route can be created efficiently.
In the tour group, the travel route on which the travel has been performed may be recorded as it is.
[0106]
FIG. 25 shows an operation explanatory diagram of a modified example of the tour group rearrangement process of one embodiment of the present invention. FIG. 25A shows a tour group before movement, and FIG. 25B shows a tour group after movement.
For example, using the tours A, C, G, and I shown in FIG. 1 → NO. 2 → NO. 3 → NO. 6 → NO. 11 → NO. 12 → NO. 13 is created and executed, the movement route executed as shown in FIG. 1 → NO. 2 → NO. 3 → NO. 6 → NO. 11 → NO. 12 → NO. 13 is inserted into the tour A where the search order is the highest, and the tours shown in FIG. The above process corresponds to the process of claim 11 of the claims.
[0107]
As described above, since the tour that has been executed is positioned at the top, when the same movement start position and movement end position are set at the next movement, the movement route is searched at once, and the movement route Can be created quickly.
The central processing unit 7 detects the capacity of the battery BAT for constant driving, and when the capacity of the battery BAT decreases to a predetermined level or lower, it is selected in step S4-10 of the route plan creation process in FIG. Change the travel route from the shortest travel route to an efficient travel route.
[0108]
As described above, when the travel speed or travel distance must be reduced, the battery status can be detected, the travel route can be changed according to the capacity of the battery, the travel route can be stopped, etc. You can prevent problems.
As described above, according to the present embodiment, the map can be easily changed even if the environment is changed. In addition, since an action plan can be established without being aware of detailed information on the environment, it is possible to perform a route plan in a more realistic time than in the prior art. Even if an accurate map cannot be prepared initially, real-world information can be easily taken in and corrected by interaction with the real world.
[0109]
【The invention's effect】
  As described above, according to claim 1 of the present invention, by detecting the surrounding environment by the environment detection means, recognizing the landmark of the movement route map composed of the landmark and the link, and sequentially moving in the direction of the link. In addition, it is possible to move between landmarks on a movement route map including landmarks and links. At this time, the travel route map is simple information of only landmarks and links, so it can be easily created and changed, and can be easily applied to any locationIn addition, an array of landmarks is generated for each link branched from the landmark where the link branches, and is stored as a database. The landmark array stored in the database matches the first landmark and the last landmark. By selecting and connecting sequentially, a series of movement paths in which landmarks are connected by links can be created. At this time, landmarks without branches are held in the database as a single array, The amount of data held in the database can be reduced, and a series of movement paths can be created by combining the arrangement of landmarks held in the database, so that the movement path can be easily searched.
[0110]
According to claim 2, the movement information of the movement route generated by the movement route generation means by having the movement information indicating the difficulty of passage of the link such as the movement distance, the movement time, and the existence probability of the link. Thus, for example, it is possible to select an efficient travel route by selecting a travel route having a short total travel distance, a travel route having a short total travel time, and a travel route having a high existence probability.
[0111]
According to claim 3, when passing through a route that can pass or cannot pass by a time such as a signal or a railroad crossing, the signal or a railroad crossing passes through a route passing through the signal or a railroad crossing, During times when traffic lights and railroad crossings are not allowed, a route that avoids traffic lights and railroad crossings can be searched, and a route with good movement efficiency such as passing through can be searched.
[0112]
According to claim 4, by sequentially following the links connected to the landmark, the landmark is a passing point, and the link is a directly movable route, so the route connecting the landmark and the link is searched. By doing so, it is possible to easily generate a route to be moved.
[0114]
  Claim5According to the above, the identification number is assigned to the landmark, and the database is configured as an array of the assigned identification numbers, so the amount of data constituting the database can be reduced, and only the identification number match or mismatch is detected. Thus, it is possible to select an arrangement that can form a series of movement paths, and to easily create a movement path.
[0115]
  Claim6According to the above, when there is a difference between the environment when actually moved and the map information, the map information is changed to match the actual environment. For example, if an obstacle that did not exist at the time of map information creation exists in the part that is the travel route in the map information, it can not move, so processing such as cutting the link in the map information is performed next time During the movement, the movement route passing through the movement route is not selected, and a movement route suitable for the actual environment can be created.
[0116]
  Also, when there are no obstacles that existed at the time of map information creation, a new landmark and link are added to that part so that it can be selected as a part of the movement route at the next movement, and the optimum movement It has the feature that a route can be created.
[0117]
  Claim7The map information can be made closer to the actual environment by changing the movement information given to the link according to the actual environment of the movement route detected by the environment detection means, and the next movement route is created. Sometimes, it has the feature that it can create a travel route that is more suitable for the actual environment.
[0118]
  Claim8When the movement information given to the link is changed, the environment detection means is reflected by reflecting the movement information previously given to the link in the change of the movement information by the environment detected by the environment detection means. Variations in the environment detected inThe injuryIt has features such as being absorbed and being able to approximate movement information to the real environment.
[0119]
  Claim9According to the present invention, the database search order is rearranged in the descending order of frequency according to the frequency used as the travel route, so that the database search time can be shortened when creating the travel route.
  Claim10According to the above, it is not necessary to newly create the actual travel route by storing the actual travel route in the database, and a new travel route is generated based on the travel route once created. Therefore, it has the feature that the creation time of the movement route can be shortened.
[Brief description of the drawings]
FIG. 1 is a block diagram of an embodiment of the present invention.
FIG. 2 is a schematic configuration diagram of map information according to an embodiment of the present invention.
FIG. 3 is an operation flowchart of map creation processing according to an embodiment of the present invention.
FIG. 4 is an operation explanatory diagram of map information generation processing according to an embodiment of the present invention.
FIG. 5 is an operation explanatory diagram when setting landmarks in map information generation processing according to an embodiment of the present invention;
FIG. 6 is an operation flowchart of database processing according to an embodiment of the present invention.
FIG. 7 is a diagram showing a state in which an identification number is assigned in the database processing according to an embodiment of the present invention.
FIG. 8 is a configuration diagram of movement information provided in the database processing according to the embodiment of this invention.
FIG. 9 is an explanatory diagram of a modified example of the movement information obtained by the database processing according to the embodiment of this invention.
FIG. 10 is a configuration diagram of a tour group generated by database processing according to an embodiment of the present invention.
FIG. 11 is an operation flowchart of movement processing according to an embodiment of the present invention.
FIG. 12 is an operation flowchart of a route plan creation process according to an embodiment of the present invention.
FIG. 13 is an operation explanatory diagram of a tour generation method according to an embodiment of the present invention.
FIG. 14 is an operation explanatory diagram of a tour generation method according to an embodiment of the present invention.
FIG. 15 is an operation flowchart of map information update processing according to an embodiment of the present invention.
FIG. 16 is an explanatory diagram of an operation when a landmark and a link are added according to an embodiment of the present invention.
FIG. 17 is an explanatory diagram of an operation when a landmark and a link are added according to an embodiment of the present invention.
FIG. 18 is an explanatory diagram of an operation when a landmark and a link are deleted according to an embodiment of the present invention.
FIG. 19 is an explanatory diagram of an operation when a landmark and a link are deleted according to an embodiment of the present invention.
FIG. 20 is an operation flowchart of information update processing according to an embodiment of the present invention.
FIG. 21 is a diagram for explaining a method of calculating the latest travel distance and travel time according to an embodiment of the present invention.
FIG. 22 is a diagram for explaining a modification of the latest moving distance and moving time calculation method according to the embodiment of the present invention.
FIG. 23 is a diagram for explaining an existence probability changing operation according to an embodiment of the present invention;
FIG. 24 is an operation explanatory diagram of tour group rearrangement processing according to an embodiment of the present invention;
FIG. 25 is an operation explanatory diagram of a modified example of the tour group rearrangement process according to the embodiment of this invention.
[Explanation of symbols]
1 Autonomous mobile robot device
2 Map information
3 Map memory
4 Environmental sensors
5 I / O devices
5a Input device
5b display
6 Interface circuit
7 Central processing unit
8 Program memory
9 Drive circuit
10 Drive mechanism

Claims (10)

A mobile device that moves in a changing environment,
A plurality of landmarks installed at movable positions in the moving environment;
A movement route map holding means for holding a movement route map composed of a link connecting between directly movable landmarks among the plurality of landmarks;
A travel route generating means for generating a travel route based on the travel route map held in the travel map holding means;
A movement control means for controlling movement based on the movement path generated by the movement path generation means; an environment detection means for detecting the surrounding environment;
The environment state detected by the environment detection means when there is a difference between the environment state detected by the environment detection means and the movement route map held by the movement route map holding means. possess a travel route map update means for updating the travel route map to reflect the map,
The movement route map holding means holds a series of movement routes set as an array of a plurality of landmarks and links as divided movement routes divided for each branch of the link,
The moving device is characterized in that the moving route generating means generates a desired moving route by sequentially connecting the divided moving routes .  The movement route map holding unit holds movement information indicating a difficulty level when moving between the landmarks connected to each link, and the movement route generation unit includes the movement information among the generated movement routes. The mobile device according to claim 1, wherein a route having a small sum is selected.  The travel route map holding unit sets the travel information according to an absolute time, and the travel route generation unit selects a travel route with a small sum of the absolute times from the generated travel routes. The moving device according to claim 1 or 2. The movement route map holding unit holds a series of movement routes as an array of the plurality of landmarks, and the movement route generation unit generates a desired movement route by combining the arrangement of landmarks. The moving device according to any one of claims 1 to 3. The movement route map holding unit holds the series of movement routes as an array of identification numbers assigned in advance to the plurality of landmarks, and the movement route generation unit generates the movement route with the array of identification numbers. 5. The movement apparatus according to claim 1, wherein the movement control unit performs movement control so as to sequentially follow the identification numbers arranged by the movement path generation unit. The movement route map update unit is configured such that when a place where movement is possible on the movement route map held in the movement route map holding unit is impossible to move as a result of detection by the environment detection unit. Deletes the landmark and the link connected to a place where the movement route map held by the movement route map holding unit becomes impossible to move, and the movement held by the movement route map holding unit If the location that is impossible to move on the route map is movable as a result of detection by the environment detection means, the movement route map held in the movement route map holding means can be moved. The moving apparatus according to claim 1, wherein the landmark and the link are added to all the places that are formed. 7. The apparatus according to claim 1, further comprising a movement information update unit configured to detect the movement information by the environment detection unit during movement and update the movement information according to a movement result after movement. Mobile device. 8. The mobile device according to claim 7, wherein the movement information update unit reflects past data when the movement information is updated. Detecting the frequency with which the series of travel routes are used as a travel route, and according to the frequency detected by the frequency detection means, the series of travel routes in the database provided in the travel route map holding unit in the order of access. It has a database changing means to reorder sequentially. The moving device according to claim 1, wherein: The moving apparatus according to claim 1, further comprising a database control unit that holds the actually moved moving route in the database as the series of moving routes.

Images