Control device

A control device,

• a processor, configured to makes output of a control target follow a prescribed target command and configured to perform
• a model prediction control, wherein the model prediction control has a prediction model in which a correlation between a prescribed state variable related to the control target and a control input to the control target is defined in a form of a prescribed state equation, the model prediction control is performed based on the prediction model on the prescribed target command to be followed by the output of the control target according to a prescribed evaluation function in a prediction interval of a prescribed time width, and a value of the control input is output at least at an initial time of the prediction interval,
• a first acquisition which acquires a position of a real obstacle that is an actual obstacle with respect to the control target, and
• a set operation which sets a position of a virtual obstacle associated with the real obstacle based on the position of the real obstacle acquired by the first acquisition in a manner that the virtual obstacle is positioned substantially symmetrically to the real obstacle with reference to a following target trajectory of the control target that is based on the prescribed target command;
• and
• a servo integrator to which a deviation between the prescribed target command and the output of the control target is input,
• wherein stage costs calculated by the prescribed evaluation function comprise:
• a state quantity cost which is a stage cost related to the prescribed state variable, a control input cost which is a stage cost associated with the control input, a first stage cost which is associated with a first probability potential field representing a probability that the real obstacle is present based on the position of the real obstacle, and a second stage cost which is associated with a second probability potential field representing a probability that the virtual obstacle is present based on the position of the virtual obstacle and having a probability value equal to or greater than that of the first probability potential field,
• the state variable associated with the control target comprises a prescribed integration term represented by a product of the deviation and a prescribed integration gain, and
• the prescribed integration gain becomes smaller as a distance between the control target and the real obstacle is reduced.