Off-Road Stuck Probability Estimation Using Road Severity Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in accurately determining the stuck probability of vehicles traveling off-road, which is crucial for preventing vehicles from getting stuck on unpaved roads.
Innovation Solution
An off-road travel assistive device that includes a receiver for collecting travel data, a road surface severity estimator using deep learning, and a stuck probability determiner employing a meta-model to comprehensively assess variables such as speed, wheel slip rate, and road surface severity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a rule-based model is used to determine stuck probability, then the determination process is simple and fast, but the accuracy is low due to complex driving habits and traveling conditions affecting the results
Solution Approach 1:
The patent transforms the stuck probability determination from a rule-based approach to a machine learning-based approach. The system collects multiple travel data parameters (accelerator open value, wheel slip rate, steering angle, gear ratio, speed, road surface type) and uses a trained machine learning model to predict stuck probability. This parameter transformation enables the system to handle complex nonlinear relationships between driving habits, traveling conditions, and stuck probability, significantly improving accuracy while maintaining real-time determination capability.
2Measurement precision
If multiple travel data variables are collected and analyzed, then the stuck probability determination accuracy is improved, but the device complexity increases
Solution Approach 1:
The patent integrates multiple existing vehicle sensors and systems into a unified stuck probability determination system. The receiver unit collects data from existing vehicle components (accelerator pedal sensor, wheel speed sensor, steering angle sensor, gear position sensor, speed sensor, road surface sensor), and the controller uses a machine learning model to process this multi-source data. This multi-functional integration improves determination accuracy without requiring entirely new hardware, thereby limiting the increase in device complexity.
Solution Approach 2:
The patent introduces a receiver unit as an intermediary component that centralizes data collection from multiple vehicle sensors. This receiver unit aggregates travel data (accelerator open value, wheel slip rate, steering angle, gear ratio, speed, road surface type) and transmits it to the controller for machine learning-based analysis. The intermediary structure simplifies the overall system architecture by providing a single data collection and processing interface, reducing complexity despite handling multiple variables.
3Reliability
If real-time stuck probability determination is implemented, then preventive measures can be taken before getting stuck, but the computational load and processing time increase
Solution Approach 1:
The patent implements preliminary action by continuously collecting and analyzing travel data in real-time to determine stuck probability before the vehicle actually gets stuck. The machine learning model processes multiple parameters (accelerator open value, wheel slip rate, steering angle, gear ratio, speed, road surface type) continuously, enabling the system to predict stuck conditions in advance and alert the driver or activate autonomous control modes preventively, rather than reacting after the vehicle is already stuck.
Solution Approach 2:
The patent replaces traditional mechanical rule-based determination systems with a machine learning-based computational system. Instead of using predefined rules and thresholds to determine stuck probability, the system employs a trained machine learning model that can process multiple travel data parameters simultaneously and provide real-time predictions. This substitution enables faster and more accurate real-time determination while reducing the computational burden of complex rule evaluation.
Data Source
AI summary
An off-road travel assistive device includes a receiver configured to receive travel data of a vehicle, and an estimator. The estimator includes a road surface severity estimator configured to estimate road surface severity data based on the travel data of the vehicle. The device also includes a stuck probability determiner configured to determine probability of a stuck state of the vehicle occurring using the travel data and the road surface severity data. The stuck probability determiner determines stuck probability using a meta-model.