MODEL PREDICTIVE CONTROL DEVICE

• Main Authors: FUJII MASAKAZU, YOSHIMITSU RYO
• Format: Patent
• Language: English; Japanese
• Published: 11.07.2019
• Subjects: CONTROL OR REGULATING SYSTEMS IN GENERAL; CONTROLLING; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS ORELEMENTS; PHYSICS; REGULATING

To suppress destabilization in the computation to solve an optimization problem of MPC by a continuous deformation method.SOLUTION: A model predictive control device 1 includes a state quantity acquisition unit 2 that acquires information on the position of a moving body 8 as a state quantity, a storage unit 3 in which a target position is stored, a constraint setting unit 4, a parameter setting unit 5, and a calculation unit 6. As a cost function at the time when an optimization problem of MPC for determining the moving path of the moving body 8 is made an unconstrained optimization problem, the calculation unit 6 sets the cost function by adding a penalty term to relax the constraint along the time axis on the horizon. Next, the calculation unit 6 solves the unconstrained optimization problem by the continuous deformation method, and sets a control amount of the first step in the obtained horizon to the control amount of the moving body 8 at the time when one step of time has elapsed in the horizon. The model predictive control device 1 gives a command 7 based on the control amount to a moving device 12 of the moving body.SELECTED DRAWING: Figure 1 【課題】 ＭＰＣの最適化問題を連続変形法で解く計算の不安定化を抑制する。【解決手段】 モデル予測制御装置１は、移動体８の位置の情報を状態量として取得する状態量取得部２と、目標位置が記憶された記憶部３と、制約設定部４と、パラメータ設定部５と、計算部６とを備える。計算部６は、移動体８の移動経路を求めるためのＭＰＣの最適化問題を、無制約化した最適化問題にするときのコスト関数として、ホライズン上で時間軸方向に沿い制約が緩和されるペナルティ項を加えたコスト関数を設定する。次いで計算部６は、無制約化された最適化問題を連続変形法で解き、得られたホライズンにおける１ステップ目の制御量を、ホライズンで１ステップ分の時間が経過した時点での移動体８の制御量とし、モデル予測制御装置１は、その制御量に基づく指令７を移動体の移動装置１２へ与える。【選択図】図１