Heavy load AGV hydraulic suspension posture keeping method

CN121224356BActive Publication Date: 2026-06-26MASCH TECH DEV CO LTD +1
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
MASCH TECH DEV CO LTD
Filing Date
2025-09-02
Publication Date
2026-06-26

AI Technical Summary

Technical Problem

In existing technologies, the attitude and wheel load control schemes of heavy-duty AGVs typically employ a single technology at the algorithm layer, which cannot effectively distinguish between static road geometry and high-frequency suspension vibration in the time domain. This results in a lag in prior knowledge of road conditions, limiting the timeliness of feedforward control, causing significant model mismatch due to instantaneous centroid shift, and neglecting factors such as oil temperature and viscosity changes at the hydraulic actuation layer, leading to deviations between actual and theoretical flow rates, which affect vehicle positioning accuracy and structural safety.

Method used

A four-source coupled road surface and centroid estimation link is adopted. The road surface curvature and load distribution are calculated in real time through a sliding mode extended Kalman filter. Combined with a model predictive controller and a graph convolutional network, the displacement and pressure commands of multiple cylinders are coordinated in real time. A hydraulic topology graph is constructed for coupled tension compensation to achieve dynamic horizontal control of the vehicle platform.

Benefits of technology

Synchronously updating road priors and load distribution within a millisecond timescale significantly improves dynamic levelness and energy utilization, solves the problem of synchronous estimation of load distribution and transient state of hydraulic network, and improves vehicle positioning accuracy and structural safety.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121224356B_ABST
    Figure CN121224356B_ABST
Patent Text Reader

Abstract

The application discloses a heavy-load AGV hydraulic suspension posture keeping method, and belongs to the technical field of heavy-load AGV control. The method obtains a vehicle-road combined state vector through a sliding mode-extended Kalman filter, obtains a vehicle linearized dynamic matrix and a covariance thereof through a gated recurrent unit network, and on the basis, uses a variable prediction time domain model prediction controller to output a target cylinder displacement instruction and a pressure instruction, and corrects the instruction in real time, so as to realize dynamic level control of a vehicle platform. The method forms a perception, decision and execution closed-loop adaptive link through a four-source coupled road and mass center estimation link, a double-layer decision architecture composed of a gated recurrent network and a model prediction control, and a hydraulic network tension compensation mechanism based on a graph convolution network, can synchronously update a road prior, a load distribution and a multi-cylinder pressure instruction in a millisecond time scale, and thus significantly improves dynamic level and energy utilization.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Leveling and stability evaluation method and system for self-adaptive terrain of truck-mounted crane supporting leg

    CN120068639A

  • Neural network driven vehicle adaptive cruise control method

    CN120096565A