Behavior estimating system
A behavior estimating system is provided. According to the system, an estimated trajectory which provides the basis on which the behavior of an agent is controlled is generated according to a second model which represents a motion of an instructor in which the position and the displacing velocity of the position of a state variable and the time differential values thereof continuously change, in addition to the position of a characteristic point of a reference trajectory which represents a motion of the instructor and a plurality of first models which represent a plurality of shape characteristics of reference trajectories. A behavior manner corresponding to a first model whose fluctuation, which is allowed under a condition that an estimated trajectory passes a characteristic state variable or a range in the vicinity thereof, is the smallest and whose stability is the highest is estimated as the behavior manner of the instructor.
1. A behavior estimating system for estimating a behavior manner of an instructor, comprising:
a state recognizing element which recognizes a position of a state variable, which is displaced as the instructor moves, at one or a plurality of reference time points as the position of one or a plurality of reference points; and
a trajectory generating element which generates, on the basis of the position or positions of the reference point or points recognized by the state recognizing element, a plurality of estimated trajectories representing estimated positions of the state variable in a time-series manner according to a plurality of first models which correspond to a plurality of behavior manners of the instructor and represent shape characteristics of individual reference trajectories indicative of the time-series changing manners of the position of the state variable and a second model which represents a motion of the instructor in which the position of the state variable and one or a plurality of n-th order time differential values (n=1, 2,. . .) thereof continuously change, under a condition that the estimated trajectories pass the reference point or points or a range or ranges in the vicinity thereof,
wherein fluctuations of the first models are allowed, stabilities of the first models in the process of generation of the plurality of estimated trajectories by the trajectory generating element are assessed on the basis of cumulative amounts of the fluctuations, and a behavior manner corresponding to one of the first models which exhibits a highest stability is estimated as a behavior manner of the instructor.
2. The behavior estimating system according to claim 1 , wherein
the trajectory generating element generates an estimated trajectory in a forward direction as the estimated trajectory by estimating the position of the state variable at each time point in the forward direction in a time-series manner, and generates an estimated trajectory in a reverse direction as the estimated trajectory by estimating the position of the state variable at each time point in the reverse direction.
3. The behavior estimating system according to claim 1 , wherein
the trajectory generating element generates a plurality of the estimated trajectories on the basis of a plurality of the first models respectively corresponding to a plurality of behavior manners respectively of the instructor which change from a first behavior manner to a second behavior manner at different time points.
4. The behavior estimating system according to claim 3 , wherein
the trajectory generating element generates a plurality of the estimated trajectories on the basis of a plurality of the first models respectively corresponding to a plurality of behavior manners respectively of the instructor which change, at different time points, from the first behavior manner for the instructor to apply a force to an object to the second behavior manner after the instructor applies the force to the object.
5. The behavior estimating system according to claim 1 , wherein
the trajectory generating element defines the shape characteristics of the reference trajectories by attractors and the positions of attracting points thereof and generates the estimated trajectories by using the first models which fluctuate by displacements of the attracting points, and
the behavior estimating system assesses the first models having smaller cumulative displacement amounts of the attracting points of the attractors as the ones having higher stabilities.