Adaptive Gear Control via Learning Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing automatic gear shifting systems for human-powered vehicles are insufficient in providing comfortable control across various riding situations and environments, as they rely on threshold-based determinations that do not adapt well to changing conditions such as riding purpose and environment.
Innovation Solution
A creation device that uses learning algorithms to create different models for producing output information concerning the control of vehicle components, such as gear stage and gear ratio, based on acquired input information including traveling speed, cadence, attitude, posture, and environmental data, allowing for adaptive control that minimizes rider discomfort.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If threshold-based determination is used for automatic gear shifting control, then the control system is simple to implement, but the control accuracy and rider comfort are insufficient
Solution Approach 1:
The patent replaces the mechanical threshold-based determination system with a machine learning model that processes sensor inputs. The learning model substitutes the simple threshold comparison mechanism with a more sophisticated computational approach that can capture complex non-linear relationships between riding conditions and optimal gear selection, thereby improving control accuracy while maintaining implementation simplicity through software-based solutions.
Solution Approach 2:
The patent changes the control parameters from fixed threshold values to dynamic predictions generated by a learning model. Instead of comparing sensor readings against predetermined thresholds, the system uses the learning model to predict optimal gear shifting timing and gear stage based on multiple input parameters including cadence, speed, torque, and riding environment, allowing the control system to adapt to varying riding conditions.
2Device complexity
If a single control model is used for all riding situations, then the device complexity is low, but the adaptability to various situations and environments is poor
Solution Approach 1:
The patent segments the control system into multiple specialized learning models, each trained for specific riding situations or environments. Instead of using one general model for all conditions, the system divides the control task into multiple focused models that can be selectively applied based on the current riding context, improving adaptability while managing complexity through modular model selection.
Solution Approach 2:
The patent introduces dynamic model selection capability where the system can switch between different learning models based on detected riding conditions. The control system dynamically determines which model to apply based on inputs such as riding purpose, terrain type, and environmental factors, allowing the system to adapt its behavior to match the current situation without requiring a single overly complex universal model.
Data Source
AI summary
A creation device includes an acquisition part that acquires input information concerning traveling of a human-powered vehicle; and a creation part that creates different learning models that each produce output information concerning control of a component of the human-powered vehicle based on input information acquired by the acquisition part.