Testing device and machine learning device
A testing device performing a worst case scenario test includes a machine learning device and the machine learning device observes data representing a test item and data representing an operation state and specifications of a manufacturing machine as state variables representing a current state of an environment. Further, the machine learning device acquires determination data representing a suitability determination result of an operation state of the manufacturing machine obtained when the test item is executed, and performs learning by using the state variables and the determination data in a state where manufacturing machine operation state data and manufacturing machine specification data are associated with test item data.
1. A testing device for testing a manufacturing machine at a development stage of the manufacturing machine to search for a test item that causes the manufacturing machine to become unstable, the testing device being configured to determine the test item with respect to an operation state of the manufacturing machine and specifications of the manufacturing machine, using at least one processor, the testing device comprising:
a machine learning device that learns determination of the test item with respect to the manufacturing machine, wherein
the machine learning device includes:
a state observation unit that observes test item data representing the test item, manufacturing machine operation state data representing the operation state of the manufacturing machine, and manufacturing machine specification data representing specifications of the manufacturing machine, as a state variable representing a current state of an environment;
a determination data acquisition unit that acquires determination data representing a suitability determination result of the operation state of the manufacturing machine, the suitability determination result being obtained when the test item is executed;
a learning unit that performs learning by using the state variable and the determination data of one or more previous cycles, such that the manufacturing machine operation state data and the manufacturing machine specification data are used to determine the test item to be performed at a next cycle; and
an inference result output unit that outputs an operation command of a test based on the test item, based on the state variable observed by the state observation unit and a learning result obtained by the learning unit.
2. The testing device according to claim 1 , wherein
the learning unit includes:
a reward calculation unit that obtains a reward related to the suitability determination result; and
a value function update unit that updates a function representing a value of the test item with respect to the operation state of the manufacturing machine and specifications of the manufacturing machine, by using the reward, wherein
the reward calculation unit calculates a higher reward as an operation of the manufacturing machine becomes more unstable.
3. The testing device according to claim 1 , wherein the learning unit calculates the state variable and the determination data in a multilayer structure.
4. A testing device for testing a manufacturing machine at a development stage of the manufacturing machine to search for a test item that causes the manufacturing machine to become unstable, using at least one processor, wherein
the testing device is configured to determine the test item with respect to an operation state of the manufacturing machine and specifications of the manufacturing machine and the testing device comprises a machine learning device that has learned determination of the test item with respect to the manufacturing machine, and
the machine learning device includes:
a state observation unit that observes test item data representing the test item, manufacturing machine operation state data representing the operation state of the manufacturing machine, and manufacturing machine specification data representing specifications of the manufacturing machine, as a state variable representing a current state of an environment;
a learning unit that has performed learning in a state where the manufacturing machine operation state data and the manufacturing machine specification data are used to determine the test item to be performed at a next cycle; and
an inference result output unit that outputs an operation command of a test based on the test item, based on the state variable observed by the state observation unit and a learning result obtained by the learning unit.
5. The testing device according to claim 1 , wherein the machine learning device exists in a cloud server.
6. The testing device according to claim 4 , wherein the machine learning device exists in a cloud server.
7. A machine learning device for testing a manufacturing machine at a development stage of the manufacturing machine to search for a test item that causes the manufacturing machine to become unstable, using at least one processor, the machine learning device comprising:
a state observation unit that observes test item data representing the test item, manufacturing machine operation state data representing an operation state of the manufacturing machine, and manufacturing machine specification data representing specifications of the manufacturing machine, as a state variable representing a current state of an environment;
a determination data acquisition unit that acquires determination data representing a suitability determination result of the operation state of the manufacturing machine, the suitability determination result being obtained when the test item is executed;
a learning unit that performs learning by using the state variable and the determination data of one or more previous cycles, such that the manufacturing machine operation state data and the manufacturing machine specification data are used to determine the test item to be performed at a next cycle; and
an inference result output unit that outputs an operation command of a test based on the test item, based on the state variable observed by the state observation unit and a learning result obtained by the learning unit.
8. A machine learning device for testing a manufacturing machine at a development stage of the manufacturing machine to search for a test item that causes the manufacturing machine to become unstable, using at least one processor, the machine learning device comprising:
a state observation unit that observes test item data representing the test item, manufacturing machine operation state data representing an operation state of the manufacturing machine, and manufacturing machine specification data representing specifications of the manufacturing machine, as a state variable representing a current state of an environment;
a learning unit that performs learning in a state where the manufacturing machine operation state data and the manufacturing machine specification data are used to determine the test item to be performed at a next cycle; and
an inference result output unit that outputs an operation command of a test based on the test item, based on the state variable observed by the state observation unit and a learning result obtained by the learning unit.