Offline ai-powered diagnostics and predictive maintenance system and method for automotive vehicles

The on-device diagnostic system addresses cloud-based limitations by processing data locally, ensuring reliable, interactive, and secure vehicle diagnostics and predictive maintenance, enhancing user empowerment and safety.

WO2026120546A1PCT designated stage Publication Date: 2026-06-11SANDLOGIC TECHNOLOGIES PVT LTD
View PDF 3 Cites 0 Cited by

Patent Information

Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
SANDLOGIC TECHNOLOGIES PVT LTD
Filing Date
2025-12-05
Publication Date
2026-06-11

AI Technical Summary

Technical Problem

Existing vehicle diagnostic systems rely on cloud-based infrastructure, which introduces connectivity limitations, latency, high operational costs, data security risks, and lack of user-friendly interaction, making them inaccessible and unreliable, especially in remote areas.

Method used

An on-device diagnostic system powered by a Large Language Model (LLM) that processes data locally, integrating sensor interfaces, microcontroller firmware, and hardware accelerators for real-time analysis, enabling interactive diagnostics and predictive maintenance without internet connectivity.

Benefits of technology

The system provides continuous, secure, and cost-effective vehicle health monitoring, empowering users with real-time fault detection and proactive maintenance, reducing downtime and enhancing safety and accessibility.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure IMGF000021_0001
    Figure IMGF000021_0001
  • Figure IMGF000021_0002
    Figure IMGF000021_0002
  • Figure IMGF000021_0003
    Figure IMGF000021_0003
Patent Text Reader

Abstract

The present invention relates to a hardware-implemented, on-device vehicular diagnostic and predictive-maintenance system configured to operate entirely offline without reliance on cloud computation. The system comprises a sensor-interface circuit for acquiring vehicular telemetry, an analog-to-digital converter for signal digitization, a diagnostic microcontroller for generating fused telemetry snapshots, a hardware arithmetic accelerator for executing quantized diagnostic computation kernels, a secure storage module for rule-tables and calibration profiles, a multi-modal user interface for rendering diagnostic outputs, and an actuator control module for issuing corrective commands. The invention performs real-time anomaly detection, rule-based safety validation, and predictive- maintenance estimation using stored deterioration patterns and time-series telemetry. All diagnostic and predictive functions execute deterministically within the embedded hardware, thereby enhancing safety, reliability, and responsiveness across diverse vehicle platforms. The invention is best understood with reference to FIG. 1, which illustrates the overall system architecture.
Need to check novelty before this filing date? Find Prior Art