Motorcycle Turn Radius Prediction for Low-Speed Maneuvering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems fail to accurately and automatically calculate the turning radius of a motorcycle, which is crucial for low-speed maneuvers, considering factors like wheelbase, turning angle, speed, and road conditions.
Innovation Solution
A method and system using machine learning to estimate the motorcycle turn radius by collecting and analyzing driving samples, environmental information, and motorcycle characteristics, generating a geometric representation of turns, and providing projected trajectories to the driver.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If automatic calculation of turning radius is implemented, then calculation accuracy is improved, but system complexity increases
Solution Approach 1:
The patent replaces manual calculation methods with an automated machine learning-based system. The system uses a trained model that automatically processes sensor data (wheelbase, turning angle, speed, road conditions) to predict turning radius, eliminating the need for complex manual calculations and improving accuracy while maintaining reasonable system complexity through software-based automation.
Solution Approach 2:
The system performs self-calibration and automatic adaptation by continuously learning from driving data. The machine learning model automatically adjusts its predictions based on collected samples without requiring manual intervention for parameter tuning, enabling the system to improve its calculation accuracy autonomously over time.
2Measurement precision
If multiple factors (wheelbase, turning angle, speed, road conditions) are considered, then prediction accuracy is improved, but data processing complexity increases
Solution Approach 1:
The patent combines multiple input factors (wheelbase, turning angle, speed, road conditions) into a unified machine learning model. The model integrates these diverse parameters simultaneously, processing them through learned relationships to produce a single turning radius prediction. This merging approach handles the complexity of multi-factor analysis while maintaining high prediction accuracy.
Solution Approach 2:
The system transforms physical parameters (wheelbase length, turning angle, speed) and environmental conditions into a format suitable for machine learning processing. The model learns optimal parameter relationships and transformations during training, automatically adjusting how input data is weighted and combined to achieve accurate predictions without manual feature engineering.
Data Source
Figure 1A
Figure 1B
Figure 1C
AI summary
A method for providing a prediction of a radius of a motorcycle turn, the method may include determining that the motorcycle is about to turn; predicting values of multiple radius of turn impacting (RTI) parameters; wherein the multiple RTI parameters are selected out of a group of parameters, wherein the selection was made during a machine learning training process, and the group of parameters comprises motorcycle kinematic parameters; determining, based on the determined values of the multiple RTI parameters, the estimated radius of the motorcycle turn; and performing a driving related operation based on the estimated radius of the motorcycle turn.