Charged particle emission device, system, method, and program

A static eliminator configured to emit charged particles to a charged object for a predetermined emission time, the static eliminator comprising:

• a pre-emission state detector configured to detect a pre-emission charged state which is a charged state of the charged object before charged particles are emitted;
• an emission time generator configured to generate an emission time of charged particles to the charged object based on a past emission time of charged particles and a charged state of the charged object after the emission;
• emission processor circuitry configured to emit charged particles to the charged object which is in the pre-emission charged state based on the emission time generated by the emission time generator;
• a post-emission state detector configured to detect a post-emission charged state which is a charged state of the charged object after the charged particles are emitted;
• machine learning processor circuitry configured to cause a predetermined machine learning model to learn a correspondence among the pre-emission charged state, the post-emission charged state, and the emission time generated by the emission time generator;
• and
• completion determination processor circuitry configured to repeatedly operate the pre-emission state detector, the emission time generator, the emission processor circuitry, the post-emission state detector, and the learning processor circuitry until the machine learning model satisfies a predetermined learning completion condition, wherein
• the learning completion condition is that an error between an emission time as output of the machine learning model and the emission time when the post-emission charged state reaches a target state or its vicinity is equal to or less than a predetermined value or a predetermined rate.

A charged particle emission system configured to emit charged particles to a charged object for a predetermined emission time, the system comprising:

• a pre-emission state detector configured to detect a pre-emission charged state which is a charged state of the charged object before charged particles are emitted;
• an emission time generator configured to generate an emission time of charged particles to the charged object based on a past emission time of charged particles and a charged state of the charged object after the emission;
• emission processor circuitry configured to emit charged particles to the charged object which is in the pre-emission charged state based on the emission time generated by the emission time generator;
• a post-emission state detector configured to detect a post-emission charged state which is a charged state of the charged object after the charged particles are emitted;
• machine learning processor circuitry configured to cause a predetermined machine learning model to learn a correspondence among the pre-emission charged state, the post-emission charged state, and the emission time generated by the emission time generator;
• and
• completion determination processor circuitry configured to repeatedly operate the pre-emission state detector, the emission time generator, the emission processor circuitry, the post-emission state detector, and the learning processor circuitry until the machine learning model satisfies a predetermined learning completion condition, wherein
• the learning completion condition is that an error between an emission time as output of the machine learning model and the emission time when the post-emission charged state reaches a target state or its vicinity is equal to or less than a predetermined value or a predetermined rate.

A charged particle emission method that emits charged particles to a charged object for a predetermined emission time, the method comprising:

• detecting a pre-emission charged state which is a charged state of the charged object before charged particles are emitted;
• generating an emission time of charged particles to the charged object based on a past emission time of charged particles and a charged state of the charged object after the emission;
• emitting charged particles to the charged object which is in the pre-emission charged state based on the emission time generated in the generating;
• detecting a post-emission charged state which is a charged state of the charged object after the charged particles are emitted;
• causing a predetermined machine learning model to learn a correspondence among the pre-emission charged state, the post-emission charged state, and the emission time generated in the generating;
• and
• repeatedly operating the detecting the pre-emission charged state, the generating, the emitting, the detecting the post-emission charged state, and the causing until the machine learning model satisfies a predetermined learning completion condition, wherein
• the learning completion condition is that an error between an emission time as output of the machine learning model and the emission time when the post-emission charged state reaches a target state or its vicinity is equal to or less than a predetermined value or a predetermined rate.

A non-transitory computer-readable medium having one or more executable instructions stored thereon causing a computer to function as a charged particle emission device configured to emit charged particles to a charged object for a predetermined emission time which, when executed by processor circuitry, cause the processor circuitry to perform a charged particle emission method that emits charged particles to a charged object for a predetermined emission time, the method comprising:

• detecting a pre-emission charged state which is a charged state of the charged object before charged particles are emitted;
• generating an emission time of charged particles to the charged object based on a past emission time of charged particles and a charged state of the charged object after the emission;
• emitting charged particles to the charged object which is in the pre-emission charged state based on the emission time generated by an emission time generator;
• detecting a post-emission charged state which is a charged state of the charged object after the charged particles are emitted;
• and
• the device comprising
• machine learning processor circuitry configured to cause a predetermined machine learning model to learn a correspondence among the pre-emission charged state, the post-emission charged state, and the emission time generated by the generating;
• and
• completion determination processor circuitry configured to repeatedly operate the detecting the pre-emission charged state, the generating the emission time, the emitting the charged particles, the detecting the post-emission charged state, and the machine learning processor circuitry until the machine learning model satisfies a predetermined learning completion condition, wherein
• the learning completion condition is that an error between an emission time as output of the machine learning model and the emission time when the post-emission charged state reaches a target state or its vicinity is equal to or less than a predetermined value or a predetermined rate.