An automatic walking method of a shovel based on an iLQR algorithm

By installing multiple types of sensors on mining loaders and combining them with the iLQR algorithm, a multi-dimensional cost function was constructed, which solved the problems of trajectory deviation and control accuracy of loaders in complex underground environments. This enabled efficient and safe automated driving control, thereby improving mine production efficiency.

CN122239699APending Publication Date: 2026-06-19NANJING MEISHAN INTELLIGENT MINING TECHNOLOGY CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
NANJING MEISHAN INTELLIGENT MINING TECHNOLOGY CO LTD
Filing Date
2025-12-10
Publication Date
2026-06-19

AI Technical Summary

Technical Problem

Existing mining loaders suffer from problems such as accumulated trajectory deviations, degraded control accuracy, insufficient real-time response, and inadequate fusion of multi-source sensor data in complex underground environments, making it difficult to achieve efficient and safe automated control.

Method used

An automatic walking control method for a loader based on the iLQR algorithm is adopted. By adding multiple types of sensors to acquire real-time status data, a nonlinear dynamic model is established, a multi-dimensional cost function is constructed, and forward simulation and backpropagation iterative calculation are performed in combination with the iLQR algorithm to realize the autonomous driving control of the loader in complex roadway environments.

Benefits of technology

It has achieved high-precision trajectory tracking, smooth driving and safe operation of underground loaders, significantly reducing operational risks and improving the efficiency of automated production in mines.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122239699A_ABST
    Figure CN122239699A_ABST
Patent Text Reader

Abstract

This invention relates to an automated walking method for a loader based on the iLQR algorithm. The method includes the following steps: Step 1, building an automated driving control system architecture for the loader; Step 2, establishing a kinematic model of the loader; Step 3, constructing a reference trajectory input model; Step 4, designing a multi-objective cost function; Step 5, iterative optimization solution based on iLQR; and Step 6, mapping and outputting control commands. This scheme achieves automated driving control of the loader during the material loading and unloading processes, thereby significantly reducing safety hazards in underground operations and improving the efficiency of automated production in mines.
Need to check novelty before this filing date? Find Prior Art