JP2004098796A - Electric power steering device - Google Patents

Abstract

The present invention relates to an electric power steering device, and can improve the active safety by suppressing the influence of side wind disturbance on steering while ensuring smooth running performance on a curved road.
A steering angle δ and a yaw rate γ of a vehicle detected by a steering angle detecting unit 22, a yaw rate detecting unit 23, a lateral acceleration detecting unit 24, and a vehicle speed detecting unit 25, respectively, based on a vehicle steering-vehicle system model. Estimating means 32 for separating and estimating a lateral gradient disturbance φ _rb and another lateral disturbance φ _cw among lateral disturbances of the vehicle from the vehicle acceleration V, the lateral acceleration _Gy, and the vehicle speed V. , The steering assist torque generated by the electric motor 11 is corrected so as to suppress the lateral disturbance φ _cw obtained by removing the lateral gradient disturbance from the lateral disturbance of the vehicle.
[Selection diagram] Fig. 1

Description

【００２４】
図２に示す関係から、オブザーバの状態方程式として次式（Ａ）が導出される。
【００２５】
【数１】

【００２６】
上式（Ａ）において、ゲインａ_１１，ａ_１２，ａ_２１，ａ_２２，ｂ_１，ｂ_２はいずれも車速に応じて決定し、ゲインｌ_１１，ｌ_１２，ｌ_２１，ｌ_２２，ｌ_３１，ｌ_３２，ｌ_４１，ｌ_４２はいずれもモータの周波数特性を考慮して数式演算により適正値を設定することができる。
したがって、オブザーバ３２では、前輪舵角δ_{ｆ ＿ ｓ}，ヨーレイトγ_ｓ，横加速度Ｇ_{ｙ ＿ ｓ}，車速Ｖの各状態量を入力量として、横速度推定値ｖ_ｅとヨーレイト推定値γ_ｅと横力外乱推定値φ_{ｃｗ ＿ ｅ}と横勾配外乱推定値φ_{ｒｂ ＿ ｅ}との各状態量を推定することができる。
【００２７】
図５に示すように、実際の操舵系は機械的な系とＤＣモータの系とで構成される。また、慣性系は、ステアリングホイール，ＤＣモータ，タイヤから成り、弾性系はトーションバーとタイヤから成る。ここで、キングピン回りに微分方程式をたてると次式（１）〜（３）のようになる。
【００２８】
【数２】

【００２９】
ここで、δ_ｆは前輪実舵角、αはハンドル角（コラム軸回り）、Ｔ_ｓは路面からのセルフアライニンクトルク、Ｔ_ｍはＤＣモータの付加トルク、Ｔ_ｈはドライバトルク（トルクセンサ値）である。なお、系のパラメータＩ_ｓ，Ｃ_ｓ，Ｎ_ｔ，Ｎ_ｍ，Ｋ_ｔをについては前記の表１に記載する。
ここで、前輪実舵角δ_ｆは、センサ値であるＴ_ｈとαとから算出可能である。
【００３０】
前式（２）を変形すると、次式（４）となる。
【００３１】
【数３】

【００３２】
さらに、ここでは、図６に示すような二輪モデルを車両モデルとする。前輪実舵角δ_ｆをシステム入力として定義すると、かかるモデルの車両系の運動方程式を次式のように記述できる。
【００３３】
【数４】

【００３４】
ここで、ｖは車両の横速度，γは車両のヨーレイト，φは横方向外乱要素である。系のパラメータＩ_ｓ，Ｃ_ｓ，Ｎ_ｔ，Ｎ_ｍ，Ｋ_ｔは前記表１に記載する。
車両の横運動とヨー運動を表す微分方程式を示す上記の式（５），（６）は、一般によく知られているものである。
横方向外乱φは、前述のように、横勾配外乱φ_ｒｂとその他の横方向外乱（横力外乱又は横風外乱）φ_ｃｗとから成る（下式参照）。
【００３５】
【数５】

【００３６】
これらの横力外乱φ_ｃｗ，横勾配外乱φ_ｒｂは式（５）に外部入力として与えられるが、直接計測することはできない。そこで、これらの横方向外乱の対策として、図７に示すように、オブザーバで外乱を推定し、コントローラ（ＥＣＵ３０内の制御指令系）ではフィードフォワードループによって外乱の影響を近似的に打ち消す制御手法をとった。計測される出力信号である横加速度センサ値は、次式に示すように車両の状態量と横勾配外乱φ_ｒｂとで表現されるが、横力外乱φ_ｃｗは含まれない。なお、コントローラ３０では横外乱以外の外乱については、車両の各状態量をフィードバックして外乱の影響を抑制している。
【００３７】
【数６】

【００３８】
上記の式（８）は、２つの横方向外乱φ_ｒｂ，φ_ｃｗを分離可能であることを示しており、式（５）〜（８）を整理すると、次の状態方程式が得られる。
【００３９】
【数７】

【００４０】
ここで、未知の状態量ｖと横方向外乱φ_ｃｗ，φ_ｒｂを同時に推定するオブザーバを設計するために、次式に示すように２つの横方向外乱φ_ｃｗ，φ_ｒｂを状態量として拡張したモデルを定義する。
【００４１】
【数８】

【００４２】
上式（１１），（１２）はｖ，γ，φ_ｃｗ，φ_ｒｂの係数行列Ａ，Ｃ（下式）にかかる階級（ｒａｎｋ［Ｃ　ＣＡ　・・・ＣＡ^ｎ−１］^Ｔ）が状態数（状態量の種類）ｎを満たすので、可観測であり、未知の状態量ｖと未知の外乱φ_ｃｗ，φ_ｒｂをオブザーバによって推定することができる。ここで、ｖ，γ，φ_ｃｗ，φ_ｒｂは推定結果であり、以下のＬは推定ゲイン行列である。
【００４３】
【数９】

【００４４】
このようなオブザーバのブロック図を模式的に示すと、図８のようになる。
４次の微分方程式を制御対象モデルとして適用すると、この４次モデルは、操舵系モデルの２つの状態変数である前輪実舵角δ_ｆと前輪実舵角速度δ_ｆ´および車両系の２つの状態変数である横速度ｖとヨーレイトγとで構成される。ここで、横方向外乱φ_ｃｗ，φ_ｒｂは外部入力として制御対象に与えられる。なお、前式（１）におけるドライバトルクはゼロと仮定する（Ｔ_ｈ＝０）。
【００４５】
【数１０】

【００４６】
そして、本制御システムは、状態フィードバックループと外乱抑制フィードフォワードループとから構成する。状態フィードバックループは、十分な系の応答性と安定性を確保する。外乱抑制フィードフォワードループは、横風外乱の定常状態における影響を抑制する。式（１４）における制御入力は、次式により定める。
【００４７】
【数１１】

【００４８】
ここで，［Ｋ_ω，Ｋ_δ，Ｋ_ｖ，Ｋ_γ］は状態フィードバックゲインである。状態フィードバックループは，主に系を安定化させることを目的とする。前輪実舵角速度δ_ｆは、モータの付加電圧と電流から計算することができる。前輪実舵角δ_ｆは前式（４）から得られる。
ｖ，φ_ｃｗ，φ_ｒｂは前記のように、外乱オブザーバから推定される。一方、外乱抑制フィードフォワードループは、横力外乱φ_ｃｗの影響を低減することを目的し、具体的には横力外乱φ_ｃｗによって生じるヨーレイトの定常値をゼロにするようにフィードフォワードゲインを定める。このフィードフォワードゲインｋ_ｆｆを定める際，式（１４）においてφ_ｒｂ＝０と仮定する。
【００４９】
以上の仮定から次の状態方程式が得られる。
【００５０】
【数１２】

【００５１】
定常状態（ｄｘ／ｄｔ＝０）において式（１８）は次式のように変形される。
【００５２】
【数１３】

【００５３】
ここで、δ_ｆｓｓ，ｖ_ｓｓ，γ_ｓｓは，φ_ｃｗに対するδ_ｆ，ｖ，γの定常値である。従って、この方程式を直接解く次式が得られる。
【００５４】
【数１４】

【００５５】
式（１７）におけるγ_ｓｓを考慮すると式（１７）は、次式のように変形される。
【００５６】
【数１５】

【００５７】
従って、フィードフォワードゲインＫ_ｆｆ（＝Ｋ_{Φ ＿ ｃｗ}）は次式から得られる。
【００５８】
【数１６】

【００５９】
本発明の一実施形態としての電動パワーステアリング装置は、上述のように構成されているので、操舵アシスト補正手段（操舵アシスト補正量演算部３４及び演算部３５）３３により、前輪舵角速度対応補正量（＝Ｋ_ωω_{ｆ ＿ ｓ}）、前輪舵角対応補正量（＝Ｋ_δδ_{ｆ ＿ ｓ}）、横速度対応補正量（＝Ｋ_ｖｖ_ｅ）、ヨーレイト対応補正量（＝Ｋ_ｒｒ_ｅ）の加算値Ｔ_{ｍ ＿ ｆｂ}（＝Ｋ_ωω_{ｆ ＿ ｓ}＋Ｋ_δδ_{ｆ ＿ ｓ}＋Ｋ_ｖｖ_ｅ＋Ｋ_ｒｒ_ｅ）により基本アシストトルクＴ_{ｍ ＿ｂａｓｅ}をフィードバック補正するとともに、横力外乱対応補正量Ｔ_{ｍ ＿ ｃｗ}（＝Ｋ_{Φ ＿ ｃｗ}φ_{ｃｗ ＿ ｅ}）により基本アシストトルクＴ_{ｍ ＿ｂａｓｅ}をフィードフォワード補正するので、横方向外乱のうちから横勾配外乱を除去した横力外乱（横風外乱等）φ_ｃｗを抑制するように電動モータを制御して操舵アシストトルクを与えるので、カーブ路のスムーズな走行性能を確保しながら、操舵への横力外乱（横風外乱等）の影響を抑制して、アクティブ・セーフティを向上させることができるようになる。
【００６０】
なお、図９は本装置にかかるシミュレーション結果を示すグラフであり、本ロバスト制御あり（フィードバック制御＋フィードフォワード制御あり）、制御なし、フィードバック制御のみ、の３つを比較して示している。この図より、横力外乱φ_ｃｗと横勾配外乱φ_ｒｂが分離して推定されていることが分かる。また、φ_ｃｗ，φ_ｒｂに対するヨーレイトの応答から制御系のロバスト性を確認することができる。フィードバックループは車両挙動を安定化させ、フィードフォワードループが横風外乱の影響だけを低減させることが分かる。
【００６１】
以上、本発明の実施形態を説明したが、本発明はかかる実施形態に限定されるものではなく、本発明の趣旨を逸脱しない範囲で種々変形して実施することができる。
例えば、各ゲインＫ_ω，Ｋ_δ，Ｋ_ｖ，Ｋ_ｒ，Ｋ_{Φ ＿ ｃｗ}については、系の安定性や操舵フィーリングを向上させることができるように、車速に応じて適宜設定することが重要である。
【００６２】
【発明の効果】
以上詳述したように、本発明の電動パワーステアリング装置によれば、車両の横方向外乱のうち、横勾配外乱と他の横方向外乱（横風外乱等）とを分離して推定して、横方向外乱のうちから横勾配外乱を除去した横方向外乱（横風外乱等）を抑制するように電動モータを制御して操舵アシストトルクを与えるので、カーブ路のスムーズな走行性能を確保しながら、操舵への横風外乱の影響を抑制して、アクティブ・セーフティを向上させることができるようになる。
【００６３】
さらに、該車両の操舵角速度を検出する操舵角速度検出手段を設け、推定手段で車両のヨーレイト及び横速度を推定するように構成し、補正手段で、操舵角速度検出手段と操舵角検出手段と車速検出手段とによりそれぞれ検出された操舵角と舵角速度と車速と、推定手段により推定されたヨーレイトと横速度とから第１補正トルクを算出し、推定手段により推定された横勾配外乱と他の横方向外乱とから第２補正トルクを算出して、該第１補正トルクと第２補正トルクとに基づいて操舵アシストトルクを補正すれば、操舵アシストを適切に行なうことができる（請求項２）。
【００６４】
また、推定手段では、オブザーバによって、操舵角検出手段とヨーレイト検出手段と横加速度検出手段と車速検出手段とによりそれぞれ検出された操舵角とヨーレイトと横加速度と車速との各状態量から、状態量である横勾配外乱と他の横方向外乱とを推定することで、確実に横勾配外乱と他の横方向外乱とを分離して推定することができる（請求項３）。
【００６５】
さらに、補正手段が、操舵角，舵角速度，ヨーレイト，横速度のそれぞれに、車速に応じてそれぞれ設定されるゲインを乗算して該第１補正トルクを算出するとともに、横勾配外乱，他の横方向外乱のそれぞれに、車速に応じてそれぞれ設定されるゲインを乗算して該第２補正トルクを算出して、該第１補正トルクと第２補正トルクとに基づいて該操舵アシストトルクを補正することにより、操舵アシストをより適切に行なうことができる（請求項４）。
【図面の簡単な説明】
【図１】本発明の一実施形態としての電動パワーステアリング装置の機能構成を示すブロック図である。
【図２】本発明の一実施形態としての電動パワーステアリング装置にかかるモータ制御系の状態量を推定するオブザーバのブロック図である。
【図３】本発明の一実施形態としての電動パワーステアリング装置を説明する模式図である。
【図４】走行する車両への横方向外乱を説明する図であり、（ａ）は走行する車両の模式的平面図、（ｂ）は走行する車両の模式的後面図である。
【図５】本発明の一実施形態にかかる操舵系モデルを示す模式図である。
【図６】本発明の一実施形態にかかる車両系モデルを示す模式図である。
【図７】本発明の一実施形態にかかるステアリング制御を説明するブロック図である。
【図８】本発明の一実施形態にかかるステアリング制御を説明するオブザーバのブロック図である。
【図９】本発明の一実施形態にかかるステアリング制御のシミュレーション結果を説明するグラフであり、（ａ）は車両のヨーレイトｒについて示し、（ｂ）は車両への横風外乱φ_ｃｗについて示し、（ｃ）は車両への横勾配外乱φ_ｒｂについて示す。
【符号の説明】
１１　ＤＣモータ
２１　前輪舵角速度検出手段（操舵角速度検出手段）
２２　前輪舵角検出手段（操舵角検出手段）
２３　ヨーレイト検出手段
２４　横加速度検出手段
２５　車速検出手段
２６　操舵トルク検出手段
３０　制御手段としての電子制御ユニット（ＥＣＵ）
３１　推定手段（オブザーバ）
３２　基本制御量設定部（電動パワステ制御演算部）
３３　補正手段（外乱抑制補正手段）
３４　操舵アシスト補正量演算部（外乱抑制制御演算部）
３５　演算部[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to an electric power steering device mounted on a vehicle and assisting steering using an electric motor.
[0002]
[Prior art]
As a power steering device for an automobile, a hydraulic power steering device has become widespread, but an electric power steering device is more advantageous for performing fine-grained steering assist control. Is being developed.
It has also been proposed that a steering assist control using such an electric power steering device can improve so-called disturbance suppression performance so that disturbance such as a cross wind applied to a vehicle does not affect a driver's steering operation. (For example, see Patent Document 1).
[0003]
In this technique, in controlling a steering reaction force (corresponding to a steering assist force), a steering angle component T1 (= gain obtained by multiplying a steering angle) of a steering reaction force based on a steering angle and damping based on a steering angular velocity are used. A component T2 (= the steering angular velocity multiplied by a gain), a first vehicle body behavior suppression component T3 based on the yaw rate of the vehicle (= a gain obtained by multiplying the yaw rate), and a second vehicle body based on the lateral G of the vehicle A behavior suppression component T4 (= gain obtained by applying a gain to the lateral G) and a road surface component T5 (= gain obtained by applying a gain to the steering reaction force) based on the steering reaction force are calculated as follows: The target steering reaction force Ts is determined by adding the components T1, T2, T3, T4, and T5 of the steering reaction force, and the output torque of the reaction force motor (corresponding to the steering assist electric motor), that is, the steering handle of the steering wheel. The steering reaction force is calculated as the target steering reaction force. To return control to s.
Ts = T1 + T3 + T4 + T5 + T2
[0004]
[Patent Document 1]
JP-A-5-105100 (items 0003, 0021, FIGS. 3 and 5).
[0005]
[Problems to be solved by the invention]
However, in the above-described conventional technology (see Patent Literature 1), it is impossible to completely separate the influence of disturbance and the vehicle behavior due to steering intended by the driver or the road structure in consideration of the traveling performance of the vehicle.
That is, the disturbance acting on the lateral movement of the vehicle includes a gravitational acceleration component due to the lateral gradient of the road surface and a centrifugal force generated at the time of turning, and a disturbance generated by other factors (for example, a cross wind or a rutted road). (Gravity acceleration component, etc.) is generally necessary for running smoothly on a curved road, but other than the latter, a lateral disturbance (lateral disturbance such as a crosswind or a rutted road, hereinafter referred to as “lateral force disturbance” or “lateral force disturbance”) Side disturbance) is unnecessary to hinder the driver's steering operation, and it is desired to eliminate this influence on steering as much as possible.
[0006]
However, in the related art, the former and the latter cannot be separated, and the gravitational acceleration component necessary for running smoothly on a curved road is also removed as a disturbance.
In particular, a lateral external force applied to a running vehicle greatly affects the behavior and attitude of the vehicle, and eventually causes a decrease in steering stability, an increase in driver's driving load, a deterioration in lane keeping performance, and the like. At this time, of the lateral external force, steering is performed only for other lateral disturbances, that is, “lateral force disturbances” such as a side wind or a rutted road, without removing the lateral inclination in consideration of the traveling performance of the vehicle as a disturbance. Want to eliminate the impact on
[0007]
The present invention has been made in view of the above-described problems, and enables a lateral force disturbance and a gravitational acceleration component or the like necessary for smoothly traveling on a curved road to be separately grasped, so that a smooth curved road can be obtained. It is an object of the present invention to provide an electric power steering device capable of suppressing the influence of a lateral force disturbance on steering and improving active safety while ensuring traveling performance.
[0008]
[Means for Solving the Problems]
Therefore, in the electric power steering apparatus according to the present invention, when assisting the steering using the electric motor mounted on the vehicle, the vehicle speed V is detected by the vehicle speed detecting means, and the steering torque detecting means is provided. detects the steering torque T _h applied to the vehicle by, for controlling the electric motor based on the detected information V, T _h from vehicle speed detecting means and the steering torque detecting means by the control means.
[0009]
In this case, furthermore, by the steering angle detecting means detects the steering angle δ of the vehicle, detecting a yaw rate γ of the vehicle by the yaw rate detecting means detects the lateral acceleration G _y of the vehicle by the lateral acceleration detecting means, The steering angle δ and the yaw rate γ of the vehicle detected by the steering angle detecting means, the yaw rate detecting means, the lateral acceleration detecting means, and the vehicle speed detecting means based on the steering-vehicle system model of the vehicle by the estimating means. From the lateral acceleration _Gy and the vehicle speed V, the lateral gradient disturbance φ _rb and the other lateral disturbance φ _cw among the lateral disturbances of the vehicle are separately estimated. Then, in the control unit, the electric motor is generated from the estimation result by the estimation unit so as to suppress the lateral disturbance φ _cw obtained by removing the lateral gradient disturbance from the lateral disturbance of the vehicle by the correction unit. The steering assist torque is corrected, and the electric motor is controlled so as to obtain the corrected steering assist torque.
[0010]
Further, a steering angular velocity detecting means for detecting the steering angular velocity ω of the vehicle is further provided, the estimating means is configured to further estimate the yaw rate γ and the lateral velocity v of the vehicle, and the correcting means is provided for the steering angular velocity A first correction torque based on the steering angle δ, the steering angular speed ω, and the vehicle speed V detected by the detection unit, the steering angle detection unit, and the vehicle speed detection unit, respectively, and the yaw rate γ and the lateral speed v estimated by the estimation unit; And a second correction torque is calculated from the lateral gradient disturbance φ _rb estimated by the estimating means and another lateral disturbance φ _cw, and the second correction torque is calculated based on the first correction torque and the second correction torque. It is also preferable to correct the steering assist torque (claim 2).
[0011]
Further, the estimating means calculates the steering angle δ, the yaw rate γ, the lateral acceleration _Gy, and the vehicle speed V detected by the steering angle detecting means, the yaw rate detecting means, the lateral acceleration detecting means, and the vehicle speed detecting means, respectively. It is also preferable that an observer for estimating a lateral gradient disturbance φ _{rb of the} vehicle, which is a state _variable, and another lateral disturbance φ _cw from each state variable is provided (claim 3).
[0012]
Further, the correction means calculates the first correction torque by multiplying each of the steering angle δ, the steering angular velocity ω, the yaw rate γ, and the lateral velocity v by a gain set according to the vehicle speed V. Each of the gradient disturbance φ _rb and the other lateral disturbance φ _cw is multiplied by a gain set according to the vehicle speed V to calculate the second correction torque, and the first correction torque and the second correction torque It is also preferable to correct the steering assist torque based on the following (claim 4).
[0013]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
1 to 9 show an electric power steering apparatus according to an embodiment of the present invention, and the description will be made with reference to these figures.
As shown in FIG. 1, the electric power steering apparatus according to the present embodiment includes an electric motor (here, a DC motor) 11 as a steering actuator, detects a steering operation state of the vehicle and a state of the vehicle, and detects these. Based on the result, a target steering assist amount is set by an ECU (electronic control unit) 30, and the DC motor 11 is controlled so that the target steering assist amount is obtained.
[0014]
As shown in FIG. 3, for example, the DC motor 11 includes a steering system (here, rack and pinion 3) including a steering wheel 1, a steering shaft 2, a rack and pinion 3, a tie rod 4, steering wheels (front wheels) 5L, 5R, and the like. ). Of course, the present device can be applied not only to a pinion type electric power steering but also to a case where a small motor is added to a hydraulic power steering or a rack assist type electric power steering.
[0015]
As the means for detecting the steering operation state, and the vehicle state, the steering wheel angular velocity sensor 21 as a steering angular steering speed detecting means for detecting (front wheel steering angular velocity) omega _{f _ s} (front wheel steering angular speed detecting means), a steering angle a steering wheel angle sensor 22 as a steering angle detecting means for detecting (front wheel steering angle) [delta] _{f _ s} (front wheel steering angle detecting means), a yaw rate sensor (yaw rate sensor as a yaw rate detecting means for detecting a yaw rate gamma _s of the vehicle ) 23, a lateral acceleration sensor 24 as a lateral acceleration detecting means for detecting a lateral acceleration G _{y _ s} of the vehicle, a vehicle speed detecting means 25 for detecting a vehicle speed V, and a steering torque applied by a driver through a steering wheel (handle). a steering torque detecting means (steering torque sensor) 26 for detecting an input steering torque) T _{h _ s} is provided.
[0016]
The ECU 30, as its functional components, the estimated basic steering assist amount calculating unit for calculating a basic steering assist amount (basic assist torque) T _{m _ base} and (power steering control arithmetic unit) 31, a lateral disturbance applied to the vehicle And a correction for calculating a steering assist correction amount (assist torque correction amount) according to a disturbance component applied to the steering system and correcting the steering assist amount (disturbance suppression control amount) accordingly. Means (disturbance suppression correction means) 33 are provided.
[0017]
In the electric power steering control arithmetic unit 31, the basic steering assist amount based on the vehicle speed V detected by the steering torque detecting means inputs the steering torque T _{h _ s} and the vehicle speed detecting means 25 detected in 26 (basic assist torque) T _{m _Base} is calculated.
In the observer 32, the front wheel steering angle δ _{f _ s} , the yaw rate γ _s , the lateral acceleration G _{y _ s} detected by the front wheel steering angle detecting means 22, the yaw rate detecting means 23, the lateral acceleration detecting means 24, and the vehicle speed detecting means 25, respectively, based on the vehicle speed V, and the lateral disturbance [specifically applied to the vehicle, the lateral force estimated disturbance value (crosswind disturbance estimated value) φ _{cw _ e,} the horizontal gradient disturbance estimate phi _{rb _ e,} lateral velocity estimate v _e , the yaw rate estimated value r _e ].
[0018]
The correction means 33, the correction assist the basic assist torque _{T m _base} adding a correction steering assist amount steering assist correction amount computing unit for calculating the (corrected assist _{torque) T m _add} (disturbance suppression control arithmetic unit) 34, a and the addition unit 35 for adding the torque _{T m _add} be provided.
Steering assist correction amount computing section 34, the estimated lateral force estimated disturbance value by the observer 32 (crosswind disturbance estimated value) φ _{cw _ e,} the horizontal gradient disturbance estimate phi _{rb _ e,} lateral velocity estimate _{v e,} yaw rate estimated value gamma _e, a front wheel steering angular speed omega _{f _ s} detected by the front wheel steering angular velocity detecting means 21, and the front wheel steering angle [delta] _{f _ s} detected by the front wheel steering angular velocity detection means 21, detected by the vehicle speed detecting means 25 was based on the vehicle speed V, the calculating the correction assist torque T _{m _add} corresponding to the disturbance component applied to the steering system.
[0019]
That is, in the steering assist compensation amount calculating section 34, a front wheel steering angular velocity corresponding correction amount by multiplying the gain K _omega to the front wheel steering angular speed omega _{f _ s} a _{(= K ω ω f _ s} ), the front wheel steering angle [delta] _{f _ s} front wheel steering angle corresponding correction amount by multiplying the gain K _[delta] a _{(= K δ δ f _ s} ), lateral velocity corresponding correction amount in the horizontal velocity estimate _{v e} is multiplied by a gain _{K v} a ₍₌ K _v v _e), yaw rate-dependent correction amount by multiplying the gain _{K r} a yaw rate estimated value gamma _e a (= _{K r} γ _e), respectively calculated, the sum of the correction amount _{T m _ fb (= K ω} ω f _ s + K δ _{δ f _ s + K v v} e + K r γ e) by correcting the feedback of the basic assist torque _{T m _base.}
[0020]
Further, the lateral force calculated lateral force disturbance corresponding correction amount _{T m _ cw (= K Φ} _ cw φ cw _ e) by multiplying the gain K _{[Phi _ cw} estimated disturbance value phi _{cw _ e,} basic by the correction amount the assist torque _{T m _base} feed forward correction. Incidentally, as shown in FIG. 1, the horizontal gradient estimated disturbance value phi _{rb _ e} gain _K Φ _{_ rb} (= minute value) by multiplying the horizontal gradient disturbance corresponding correction amount _{T m _ rb (= K Φ} _ rb φ _{rb _ e)} is calculated, and this by the correction amount the basic assist torque _{T m _base} may feed forward correction, but basically, in accordance with only the next power disturbance estimation value phi _{cw _ e,} transverse as the influence of the disturbance is eliminated, an the basic assist torque T _{m _base} important to feed forward correction.
[0021]
Therefore, the control system is configured to suppress only the influence of the lateral force disturbance among the lateral gradient disturbance and the lateral force disturbance (lateral wind disturbance) separated by the estimation of the observer 32.
That is, as shown in FIG. 4, generally, the lateral gradient of the road on the curved road is provided to cancel the centrifugal force generated by bending the curve and to facilitate the turning on the curve. The lateral gradient on a straight road has a small value of about 2 to 3%, but is provided for drainage. The observer 32 configures a control system such that such a lateral gradient disturbance out of the lateral disturbances is excluded from disturbance suppression targets, and only the influence of other lateral disturbances (lateral force disturbance or side wind disturbance) is suppressed. It is doing.
[0022]
The estimation by the observer 32 will be further described. The relationship between the variables regarding the DC motor 11 can be modeled in a control block as shown in FIG.
The parameters in FIGS. 1 and 2 are shown in Table 1 below.
[0023]
[Table 1]

[0024]
From the relationship shown in FIG. 2, the following equation (A) is derived as an observer state equation.
[0025]
(Equation 1)

[0026]
In the above formula (A), the gain _{a 11, a 12, a 21} , a 22, b 1, b 2 are both determined according to the vehicle speed, gain _{l 11, l 12, l 21} , l 22, l 31 , L ₃₂ , l ₄₁ , and l ₄₂ can be set to appropriate values by mathematical calculations in consideration of the frequency characteristics of the motor.
Thus, the observer 32, the front wheel steering angle [delta] _{f _ s,} yaw rate gamma _s, the lateral acceleration _{G y _ s,} as an input variable to each state quantity of the vehicle speed V, the lateral velocity estimated value _{v e} and yaw rate estimated value gamma _e and horizontal it is possible to estimate the state quantities of the Chikaragairan estimate phi _{cw _ e} and a transverse gradient disturbance estimate φ _{rb _ e.}
[0027]
As shown in FIG. 5, the actual steering system is composed of a mechanical system and a DC motor system. Further, the inertial system includes a steering wheel, a DC motor, and a tire, and the elastic system includes a torsion bar and a tire. Here, when a differential equation is established around the kingpin, the following equations (1) to (3) are obtained.
[0028]
(Equation 2)

[0029]
Here, [delta] _f the front wheel actual steering angle, alpha is the steering wheel angle (the column axis), T _s is the self-Alai Schimmelpenninck torque from the road surface, T _m is the additional torque of the DC motor, T _h is the driver torque (torque sensor value ). The system parameters I _s , C _s , N _t , N _m , and K _t are described in Table 1 above.
Here, the front wheel actual steering angle [delta] _f, can be calculated from T _h and α as a sensor value.
[0030]
By transforming the previous equation (2), the following equation (4) is obtained.
[0031]
[Equation 3]

[0032]
Further, here, a two-wheel model as shown in FIG. 6 is used as a vehicle model. When defining the front wheel actual steering angle [delta] _f as a system input, it can describe the motion equation of the vehicle system such models as:.
[0033]
(Equation 4)

[0034]
Here, v is the lateral speed of the vehicle, γ is the yaw rate of the vehicle, and φ is the lateral disturbance factor. The system parameters I _s , C _s , N _t , N _m , and K _t are described in Table 1 above.
The above equations (5) and (6) representing differential equations representing the lateral motion and the yaw motion of the vehicle are generally well known.
As described above, the lateral disturbance φ includes a lateral gradient disturbance φ _rb and another lateral disturbance (lateral force disturbance or cross wind disturbance) φ _cw (see the following equation).
[0035]
(Equation 5)

[0036]
These lateral force disturbance φ _cw and lateral gradient disturbance φ _rb are given as external inputs in equation (5), but cannot be directly measured. Therefore, as a countermeasure against these lateral disturbances, as shown in FIG. 7, a control method of estimating the disturbance with an observer and a controller (a control command system in the ECU 30) approximately canceling out the influence of the disturbance by a feedforward loop is used. I took it. The lateral acceleration sensor value, which is an output signal to be measured, is expressed by the state quantity of the vehicle and the lateral gradient disturbance φ _rb as shown in the following equation, but does not include the lateral force disturbance φ _cw . Note that the controller 30 feeds back each state quantity of the vehicle for disturbances other than the lateral disturbance to suppress the influence of the disturbance.
[0037]
(Equation 6)

[0038]
The above equation (8) indicates that the two lateral disturbances φ _rb and φ _cw can be separated, and by rearranging the equations (5) to (8), the following state equation is obtained.
[0039]
(Equation 7)

[0040]
Here, in order to design an observer that simultaneously estimates the unknown state quantity v and the lateral disturbances φ _cw , φ _rb , the two lateral disturbances φ _cw , φ _rb are extended as state quantities as shown in the following equation. Define the model.
[0041]
(Equation 8)

[0042]
In the above equations (11) and (12), the class (rank [C CA... CA ^n-1 ] ^T ) of the coefficient matrixes A and C (the following equation) of v, γ, φ _cw , and φ _rb is the number of states. Since (type of state quantity) n is satisfied, it is observable, and the unknown state quantity v and unknown disturbances φ _cw and φ _rb can be estimated by the observer. Here, v, γ, φ _cw , φ _rb are estimation results, and L below is an estimation gain matrix.
[0043]
(Equation 9)

[0044]
FIG. 8 is a schematic block diagram of such an observer.
When the fourth-order differential equation is applied as a controlled object model, the fourth-order model, two states of actual front wheel actual steering angle [delta] _f and the front wheel is a two state variables steering angular velocity [delta] _{f 'and} the vehicle system of the steering system model It is composed of the lateral velocity v and the yaw rate γ, which are variables. Here, the lateral disturbances φ _cw and φ _rb are given to the control target as external inputs. Note that the driver torque in the above equation (1) is assumed to be zero ( _Th = 0).
[0045]
(Equation 10)

[0046]
The control system includes a state feedback loop and a disturbance suppression feedforward loop. The state feedback loop ensures sufficient system responsiveness and stability. The disturbance suppression feedforward loop suppresses the influence of the crosswind disturbance in a steady state. The control input in the equation (14) is determined by the following equation.
[0047]
[Equation 11]

[0048]
Here, [ _Kω , _Kδ , _Kv , _Kγ ] is a state feedback gain. The purpose of the state feedback loop is mainly to stabilize the system. Front wheel actual steering angular velocity [delta] _f can be calculated from the additional voltage and current of the motor. Front wheel actual steering angle [delta] _f is obtained from Equation (4).
v, φ _cw and φ _rb are estimated from the disturbance observer as described above. On the other hand, the disturbance suppression feedforward loop aims at reducing the influence of the lateral force disturbance φ _cw , and specifically, determines the feedforward gain so that the steady value of the yaw rate generated by the lateral force disturbance φ _cw is set to zero. . When determining the feedforward gain k _ff , it is assumed that φ _rb = 0 in equation (14).
[0049]
From the above assumptions, the following equation of state is obtained.
[0050]
(Equation 12)

[0051]
In the steady state (dx / dt = 0), equation (18) is modified as follows.
[0052]
(Equation 13)

[0053]
Here, δ _fss , v _ss , γ _ss are steady values of δ _f , v, γ with respect to φ _cw . Therefore, the following equation that directly solves this equation is obtained.
[0054]
[Equation 14]

[0055]
Considering γ _ss in equation (17), equation (17) is modified as follows.
[0056]
[Equation 15]

[0057]
Thus, feedforward gain _{K ff (= K Φ _ cw} ) is obtained from the following equation.
[0058]
(Equation 16)

[0059]
Since the electric power steering apparatus according to one embodiment of the present invention is configured as described above, the steering assist correction means (the steering assist correction amount calculation units 34 and 35) 33 corrects the front wheel steering angular velocity corresponding correction amount. _{(= K ω ω f _ s} ), front wheel steering angle corresponding correction amount _{(= K δ δ f _ s} ), lateral velocity corresponding correction amount ₍₌ K _{v v} e), the yaw rate corresponding correction amount ₍₌ K _{r r} e) the additional value _{T m _ fb (= K ω} ω f _ s + K δ δ f _ s + K v v e + K r r e) the basic assist torque _{T m _base} well as the feedback correction, the lateral force disturbance corresponding correction amount T since the _{m _ cw (= K Φ _} cw φ cw _ e) the basic assist torque _{T m _base} feed forward correction, the lateral force disturbance (lateral removal of the horizontal gradient disturbance among the lateral disturbance Since by controlling the electric motor so as to suppress the disturbance) phi _cw providing a steering assist torque, while ensuring smooth running performance of the curved road, suppress the influence of the transverse force disturbance to the steering (crosswind disturbance) As a result, active safety can be improved.
[0060]
FIG. 9 is a graph showing a simulation result of the present apparatus, and shows three comparisons of the present robust control (with feedback control + feedforward control), no control, and only feedback control. From this figure, it can be seen that the lateral force disturbance φ _cw and the lateral gradient disturbance φ _rb are estimated separately. Further, the robustness of the control system can be confirmed from the response of the yaw rate to φ _cw and φ _rb . It can be seen that the feedback loop stabilizes the vehicle behavior and the feedforward loop reduces only the effects of crosswind disturbances.
[0061]
Although the embodiments of the present invention have been described above, the present invention is not limited to such embodiments, and can be variously modified and implemented without departing from the spirit of the present invention.
For example, the gains _{K ω, K δ, K v} , for _{K r, K Φ _ cw,} as it is possible to improve the stability and steering feeling of the system, important to appropriately set according to the vehicle speed It is.
[0062]
【The invention's effect】
As described above in detail, according to the electric power steering apparatus of the present invention, of the lateral disturbances of the vehicle, the lateral gradient disturbance and other lateral disturbances (such as lateral wind disturbance) are estimated separately. The electric motor is controlled to suppress lateral disturbance (lateral wind disturbance, etc.) from which lateral gradient disturbance is removed from the directional disturbance, and steering assist torque is applied. Therefore, steering is performed while ensuring smooth running performance on a curved road. It is possible to improve the active safety by suppressing the influence of crosswind disturbance on the vehicle.
[0063]
Further, a steering angular velocity detecting means for detecting the steering angular velocity of the vehicle is provided, the estimating means is configured to estimate the yaw rate and the lateral speed of the vehicle, and the correcting means is used for the steering angular velocity detecting means, the steering angle detecting means, and the vehicle speed detecting means. Means for calculating a first correction torque from the steering angle, the steering angular velocity and the vehicle speed detected by the means, and the yaw rate and the lateral speed estimated by the estimating means. If the second correction torque is calculated from the disturbance and the steering assist torque is corrected based on the first correction torque and the second correction torque, the steering assist can be appropriately performed (claim 2).
[0064]
In the estimation means, the state quantity is calculated from the state quantities of the steering angle, the yaw rate, the lateral acceleration, and the vehicle speed detected by the observer by the steering angle detection means, the yaw rate detection means, the lateral acceleration detection means, and the vehicle speed detection means, respectively. By estimating the lateral gradient disturbance and other lateral disturbances, it is possible to reliably estimate the lateral gradient disturbance and other lateral disturbances separately (claim 3).
[0065]
Further, the correction means calculates the first correction torque by multiplying each of the steering angle, the steering angular velocity, the yaw rate, and the lateral speed by a gain set according to the vehicle speed, and further corrects a lateral gradient disturbance and other lateral disturbances. Each of the directional disturbances is multiplied by a gain set according to the vehicle speed to calculate the second correction torque, and the steering assist torque is corrected based on the first correction torque and the second correction torque. Thus, steering assist can be performed more appropriately (claim 4).
[Brief description of the drawings]
FIG. 1 is a block diagram showing a functional configuration of an electric power steering device as one embodiment of the present invention.
FIG. 2 is a block diagram of an observer for estimating a state quantity of a motor control system according to the electric power steering apparatus as one embodiment of the present invention.
FIG. 3 is a schematic diagram illustrating an electric power steering device as one embodiment of the present invention.
4A and 4B are diagrams for explaining a lateral disturbance to a traveling vehicle, wherein FIG. 4A is a schematic plan view of the traveling vehicle, and FIG. 4B is a schematic rear view of the traveling vehicle.
FIG. 5 is a schematic diagram showing a steering system model according to an embodiment of the present invention.
FIG. 6 is a schematic diagram showing a vehicle system model according to one embodiment of the present invention.
FIG. 7 is a block diagram illustrating steering control according to an embodiment of the present invention.
FIG. 8 is a block diagram of an observer illustrating steering control according to an embodiment of the present invention.
FIGS. 9A and 9B are graphs illustrating simulation results of steering control according to an embodiment of the present invention. FIG. 9A shows a yaw rate r of the vehicle, FIG. 9B shows a cross wind disturbance φ _cw to the vehicle, c) shows the lateral gradient disturbance φ _rb to the vehicle.
[Explanation of symbols]
11 DC motor 21 Front wheel steering angular velocity detecting means (steering angular velocity detecting means)
22 Front wheel steering angle detecting means (steering angle detecting means)
23 Yaw rate detection means 24 Lateral acceleration detection means 25 Vehicle speed detection means 26 Steering torque detection means 30 Electronic control unit (ECU) as control means
31 Estimation means (observer)
32 Basic control amount setting unit (Electric power steering control calculation unit)
33 correction means (disturbance suppression correction means)
34 Steering assist correction amount calculation unit (disturbance suppression control calculation unit)
35 Calculation unit

Claims (4)

An electric motor mounted on the vehicle, vehicle speed detecting means for detecting a vehicle speed of the vehicle, steering torque detecting means for detecting a steering torque applied to the vehicle, and detection information from the vehicle speed detecting means and the steering torque detecting means An electric power steering apparatus comprising: a control unit for controlling the electric motor based on the electric motor, and generating a steering assist torque using the electric motor,
Steering angle detecting means for detecting a steering angle of the vehicle;
Yaw rate detection means for detecting the yaw rate of the vehicle,
Lateral acceleration detecting means for detecting a lateral acceleration of the vehicle;
Based on the steering-vehicle system model of the vehicle, the steering angle, yaw rate, lateral acceleration, and vehicle speed of the vehicle detected by the steering angle detecting means, the yaw rate detecting means, the lateral acceleration detecting means, and the vehicle speed detecting means, respectively. And estimating means for separating and estimating the lateral gradient disturbance and other lateral disturbances among the lateral disturbances of the vehicle,
The control means corrects the steering assist torque generated by the electric motor based on the estimation result by the estimation means so as to suppress the lateral disturbance in which the lateral gradient disturbance is removed from the lateral disturbance of the vehicle. An electric power steering device, comprising: a correction unit that performs correction. Further comprising a steering angular velocity detecting means for detecting a steering angular velocity of the vehicle;
The estimating means is configured to further estimate the yaw rate and the lateral speed of the vehicle,
The correcting means includes a steering angle, a steering angular velocity, and a vehicle speed detected by the steering angular velocity detecting means, the steering angle detecting means, and the vehicle speed detecting means, respectively, and a yaw rate and a lateral speed estimated by the estimating means. (1) calculating a correction torque, calculating a second correction torque from the lateral gradient disturbance estimated by the estimating means and another lateral disturbance, and performing the steering based on the first correction torque and the second correction torque. The electric power steering apparatus according to claim 1, wherein the assist torque is corrected. The estimating means calculates a state quantity from the state quantities of the steering angle, the yaw rate, the lateral acceleration, and the vehicle speed detected by the steering angle detecting means, the yaw rate detecting means, the lateral acceleration detecting means, and the vehicle speed detecting means, respectively. 3. The electric power steering apparatus according to claim 1, further comprising an observer for estimating a lateral gradient disturbance and another lateral disturbance of the vehicle. The correction means calculates the first correction torque by multiplying each of the steering angle, the steering angular velocity, the yaw rate, and the lateral speed by a gain set according to the vehicle speed, and calculates a lateral gradient disturbance and other lateral directions. Multiplying each of the disturbances by a gain set according to the vehicle speed to calculate the second correction torque, and correcting the steering assist torque based on the first correction torque and the second correction torque; The electric power steering device according to any one of claims 1 to 3, characterized in that:

Images