An internet of things control method and system based on natural language processing

By acquiring user intent and entities through a natural language processing system, generating tool call sequences and performing parameter conversion, the problem of IoT devices struggling to understand complex instructions and confusing tool calls is solved, thus achieving accuracy and coordination in device control.

CN122364402APending Publication Date: 2026-07-10CHONGQING SELIS PHOENIX INTELLIGENT INNOVATION TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN Β· China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-05-08
Publication Date
2026-07-10

AI Technical Summary

Technical Problem

Existing IoT devices struggle to accurately understand complex natural language commands from users and cannot automatically coordinate multiple tools or devices to complete complex tasks, leading to deviations in control command execution and confusion in tool invocation.

Method used

The system obtains user control commands through a natural language processing system, performs intent recognition and entity recognition, generates tool call sequences, sorts and converts parameters based on a preset business logic chain, and generates CAN message data to achieve device control.

Benefits of technology

It enables efficient parsing of user-generated spoken commands, ensuring the accuracy and stability of device control, solving the problems of chaotic tool calls and disordered resource calls, and improving the convenience of human-computer interaction and the efficiency of device collaborative execution.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122364402A_ABST
    Figure CN122364402A_ABST
Patent Text Reader

Abstract

This application provides an IoT control method and system based on natural language processing. The method involves acquiring user control commands; performing intent recognition on the natural language information to determine intent data, which includes the target intent and associated entities; selecting from a preset tool list based on the target intent to generate a target tool set; sorting the target tool set based on a preset business logic chain matching the target intent to generate a tool call sequence; and responding to the tool call sequence by converting file parameters corresponding to the target intent and entities to generate message data, enabling the target tools to respond according to the message data and complete the control operation. This application solves the problems of chaotic tool calls and incorrect call order under multiple control targets by using the tool call sequence; it automatically matches parameters and fills in the generated message data by combining the target intent and corresponding entities, achieving seamless integration between upper-layer natural language interaction and lower-layer hardware control, ensuring efficient and stable execution.
Need to check novelty before this filing date? Find Prior Art