Road Surface Prediction Using Fleet and Vehicle Sensor Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle control systems lack reliable information about upcoming road surface operating conditions, leading to inadequate planning of control actions and potential safety issues or under-utilization of control capacity.
Innovation Solution
A computer system that predicts road surface operating conditions by receiving a shared operating conditions model from a remote statistics processor and combining it with real-time sensor measurement data from the vehicle to determine a vehicle-specific model, which is then used to predict upcoming operating conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If currently measured parameter values are used as estimation of upcoming parameter values, then the system is simple to operate, but the reliability of vehicle control is poor
Solution Approach 1:
The system performs preliminary actions by collecting and processing sensor data from multiple vehicles to predict upcoming road surface conditions before the vehicle encounters them. The prediction system uses historical and real-time data from fleet vehicles to proactively determine future operating conditions, allowing the control system to prepare appropriate control actions in advance rather than reacting to current conditions only.
Solution Approach 2:
The patent introduces a prediction system as an intermediary between raw sensor data and vehicle control decisions. This intermediary component processes and interprets data from multiple sources (sensor measurements, shared fleet data, environmental conditions) to generate predicted upcoming operating conditions, which then inform the vehicle control system. This intermediary layer adds complexity but significantly improves reliability by providing more accurate forward-looking information.
2Measurement precision
If a shared operating conditions model from fleet vehicles is combined with real-time sensor data, then the measurement precision of upcoming conditions is improved, but the device complexity increases
Solution Approach 1:
The system merges two distinct data sources: shared operating conditions models from fleet vehicles and real-time sensor measurements from the individual vehicle. By combining these complementary data sources, the system achieves higher measurement precision for predicting upcoming road surface conditions than either source could provide alone. The shared model provides aggregated fleet knowledge while real-time sensors provide vehicle-specific current conditions.
Solution Approach 2:
The prediction system serves multiple functions: it processes shared fleet data, integrates real-time sensor inputs, predicts upcoming conditions, and provides outputs for vehicle control. This multi-functional system handles diverse data types and performs multiple processing tasks, which increases device complexity but enables comprehensive and accurate condition prediction that benefits overall vehicle control.
3Productivity
If prediction of upcoming operating conditions is implemented, then the productivity of vehicle control planning is improved, but the loss of computational resources increases
Solution Approach 1:
The system performs computational work in advance by continuously building and updating the shared operating conditions model using fleet data. This preliminary processing of data during periods when the individual vehicle is not actively predicting reduces the real-time computational burden. The pre-processed shared model is then combined with real-time sensor data to generate predictions, improving the efficiency of control planning while managing energy consumption through distributed preprocessing.
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
First and second methods for prediction of road surface operating conditions of a vehicle are disclosed. The first method comprises receiving at least a portion of a shared operating conditions model from a remote statistics processor, receiving real-time sensor measurement data from sensors of the vehicle, determining a vehicle-specific operating conditions model based on the portion of the shared operating conditions model and the real-time sensor measurement data, and using the vehicle-specific operating conditions model to predict upcoming operating conditions for the vehicle. The second method comprises receiving respective sensor measurement data from a plurality of vehicles, determining a shared operating conditions model based on the measurement data, and causing provision of at least a portion of the shared operating conditions model to a specific vehicle. Corresponding computer systems, vehicle, server node, computer program product, and non-transitory computer-readable storage medium are also disclosed.