AI Vehicle Weight Estimation from On-Board Driving Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle load identification mechanisms rely on costly hardware devices with specialized sensors, requiring installation and maintenance, and suffer from data transmission and hardware failure issues, making them inefficient and unreliable for real-time load weight estimation.
Innovation Solution
A method using artificial intelligence models to predict vehicle weight based on low-frequency driving data from existing on-board sensors, employing up-sampling, filtering, and data fusion to generate accurate weight estimates without additional hardware, reducing costs and errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If specialized hardware sensors and load identification devices are installed in the vehicle, then measurement precision of vehicle weight is improved, but device complexity and cost increase
Solution Approach 1:
The patent replaces mechanical/physical sensing systems (load cells, specialized sensors) with an information-processing system that uses existing vehicle sensors (accelerometers, gyroscopes, tire pressure sensors) combined with AI algorithms to estimate vehicle weight. This substitution eliminates the need for dedicated weight measurement hardware while achieving accurate weight estimation through data processing and machine learning models.
Solution Approach 2:
The patent makes existing multi-functional vehicle sensors serve the additional purpose of weight estimation. Sensors originally designed for other functions (acceleration monitoring, stability control, tire pressure monitoring) are repurposed to provide data for weight calculation, eliminating the need for specialized single-function weight sensors.
2Measurement precision
If specialized sensors and hardware devices are installed for load identification, then measurement precision is improved, but ease of manufacture and installation deteriorates
Solution Approach 1:
The system utilizes vehicle components and sensors that already exist and serve themselves for multiple purposes. The existing sensor network, installed for other vehicle functions, automatically provides data for weight estimation without requiring separate installation of dedicated weight sensing hardware.
Solution Approach 2:
Existing vehicle sensors are designed to perform multiple functions simultaneously. The accelerometer serves both collision detection and weight estimation; the gyroscope serves both stability control and weight calculation; tire pressure sensors serve both safety monitoring and weight determination.
3Measurement precision
If hardware load identification devices are deployed, then measurement precision is improved, but loss of energy increases due to transmission costs
Solution Approach 1:
The patent merges the weight estimation function with the existing vehicle data processing system. Instead of separate hardware devices that independently measure and transmit weight data, the system combines weight calculation with the vehicle's existing sensor data processing pipeline, eliminating redundant transmission infrastructure.
Solution Approach 2:
The patent replaces physical transmission hardware with computational processing. Weight information is derived through AI algorithms processing existing sensor data streams, eliminating the need for dedicated weight sensor transmission channels and reducing overall data transmission requirements.
4Measurement precision
If specialized load identification hardware is installed, then measurement precision is improved, but reliability deteriorates due to hardware failure and incorrect data
Solution Approach 1:
The system implements feedback mechanisms where the AI model continuously learns from and adjusts to actual vehicle operating conditions. By comparing estimated weight with expected values based on vehicle dynamics and sensor patterns, the system can detect and correct anomalies, ensuring reliable weight estimation even when individual sensors experience temporary failures or anomalies.
Solution Approach 2:
The AI-based system performs self-validation and error detection by analyzing the consistency and plausibility of sensor data patterns. The system can identify when sensor readings are anomalous or when environmental conditions affect measurements, automatically adjusting calculations to maintain reliable weight estimation without external calibration hardware.
Data Source
AI summary
Mechanisms are provided for automatically predicting vehicle weight of a moving vehicle. Vehicle operation data is obtained from one or more on-board sensors/systems of the vehicle and features are extracted. The features are filtered based on a required working condition of the vehicle to identify intervals of features having valid feature data to thereby generate a filtered valid data. The filtered valid data is processed by a first artificial intelligence (AI) computer model based on time slices of the filtered valid data to generate a first prediction of vehicle weight, and by a second AI computer model based on buckets of key variables to generate a second prediction of vehicle weight. The first prediction is fused with the second prediction to generate a final prediction of vehicle weight which is output to downstream computing logic.