Information processing device and information processing method

Abstract

The present technology relates to an image processing device and an image processing method that make it easier to ascertain the skill learning status of a robot. This information processing device comprises: an analysis unit that analyzes skill information including a skill feature amount which is the feature amount of a skill based on an internal state of a control model that controls execution of the skill of a robot; and an output control unit that controls output of the analysis result of the skill information. The present technology is applicable to, for example, a system for executing robot skill learning.

Claims

1. An information processing device comprising: an analysis unit that analyzes skill information including skill features, which are skill features based on the internal state of a control model that controls the execution of a robot's skills; and an output control unit that controls the output of the analysis results of the skill information.

2. The information processing device according to claim 1, wherein the analysis unit generates a skill map showing the distribution of learned skills based on the skill features.

3. The information processing device according to claim 2, wherein the skill information includes at least one of the following: physicality information relating to the physicality of the robot, target object information relating to the target object which is an object used in the skill, environmental information relating to the environment surrounding the robot when the skill is executed, success rate information relating to the success rate of the skill, and a skill label indicating the attributes of the skill, and the analysis unit superimposes at least one of the physicality information, target object information, environmental information, success rate information, and skill label onto the skill map.

4. The information processing device according to claim 3, wherein the analysis unit superimposes the trajectory of the skill feature quantity of the taught skill, which is a skill taught to the robot, onto the skill map.

5. The information processing apparatus according to claim 4, wherein the analysis unit predicts the trend of the success rate of the teaching skill based on the trajectory of the skill features of the teaching skill.

6. The information processing device according to claim 5, further comprising a control unit that controls the execution of the robot's skills using the control model, wherein the control unit adjusts control parameters for controlling the execution of the teaching skills based on the trend of the predicted success rate of the teaching skills.

7. The information processing apparatus according to claim 5, wherein the analysis unit extracts the portion of the teaching skills whose predicted success rate is below a predetermined threshold as skills that require learning.

8. The information processing device according to claim 5, wherein the analysis unit predicts at least one of the robot's physicality, the target object, and the environment suitable for the teaching skill based on the skill map.

9. The information processing device according to claim 3, wherein the analysis unit predicts at least one of the robot's physicality, the target object, and the environment that is suitable for the skills where learning is lacking, based on the skill map.

10. The information processing device according to claim 2, wherein the analysis unit extracts skills that the robot can perform based on the skill map.

11. The information processing device according to claim 2, wherein the analysis unit detects the range of skills in which learning is insufficient based on the skill map.

12. The information processing apparatus according to claim 11, further comprising a learning unit that performs learning of the control model, wherein the learning unit performs relearning of the control model to adapt to the skills that are lacking in the learning.

13. The information processing apparatus according to claim 1, wherein the analysis unit determines whether or not the taught skill is executable by comparing the skill features of the learned skill with the skill features of the taught skill which is a skill taught to the robot.

14. The information processing apparatus according to claim 1, wherein the output control unit controls the display of an analysis screen showing the analysis results of the skill information.

15. The information processing apparatus according to claim 1, wherein the control model is composed of a neural network, and internal state data indicating the internal state of the control model is output from an intermediate layer of the neural network.

16. The information processing apparatus according to claim 15, wherein the internal state data is output from the lowest dimensional layer of the neural network.

17. The information processing apparatus according to claim 1, further comprising: a control unit that controls the execution of the robot's skills using the control model and outputs internal state data indicating the internal state of the control model; and a feature extraction unit that extracts the skill features based on the internal state data.

18. The information processing device according to claim 1, wherein the control model receives input information including observation information and outputs control command information indicating a control command for the robot.

19. The information processing apparatus according to claim 1, further comprising a learning unit for performing the learning of the control model.

20. An information processing method comprising: analyzing skill information, which includes skill features, which are skill features based on the internal state of a control model that controls the execution of a robot's skills; and controlling the output of the analysis results of the skill information.

Images