Electric Bicycle Sensor Verification via Predictive Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electric bicycle control systems face reliability issues due to sensor malfunctions or data unavailability, particularly with acceleration and gyroscope data, which can lead to inadequate control of the electric motor and compromised safety features.
Innovation Solution
The electric bicycle employs a bicycle bus system for communication between components, utilizing operational parameters from various ECUs, including optional acceleration and gyroscope data, to verify sensor data plausibility and control the electric motor based on more reliable parameters, using pre-trained models like decision trees to predict and verify operational parameters, even if sensor units are missing or malfunctioning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensor data from acceleration sensors and gyroscopes is used for control, then control precision and safety are improved, but system reliability deteriorates due to sensor malfunctions or data unavailability
Solution Approach 1:
The system changes the source of operational parameters by using pre-trained models to predict parameters (acceleration, angular rate, brake status, gear status) instead of relying solely on physical sensors. This allows the system to maintain control functionality even when sensor data is unavailable or malfunctioning, resolving the contradiction between measurement precision and system reliability.
Solution Approach 2:
Pre-trained models act as intermediaries between the physical sensors and the control system. When sensors are unavailable, the models predict the required parameters based on historical data and patterns, serving as a mediator that maintains system reliability without compromising control precision.
2Reliability
If multiple sensors and complex verification systems are added to improve reliability, then system reliability is improved, but device complexity increases
Solution Approach 1:
The system uses pre-trained models that can autonomously predict operational parameters without requiring additional physical sensors or complex verification hardware. The models self-service the system by providing predicted parameters directly when sensor data is unavailable, improving reliability while avoiding increased device complexity.
Solution Approach 2:
The patent replaces mechanical sensor systems with software-based pre-trained models. Instead of adding more physical sensors and complex verification hardware, the system uses machine learning models to predict parameters, substituting mechanical complexity with computational intelligence that achieves the same reliability improvement.
3Device complexity
If optional sensor units are used to reduce system complexity, then device complexity is reduced, but measurement precision deteriorates when sensors are missing
Solution Approach 1:
The system performs preliminary action by pre-training models offline with extensive data before deployment. This preliminary training enables the models to accurately predict operational parameters even when physical sensors are absent, maintaining measurement precision while allowing for reduced device complexity through optional sensor configurations.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enhances the reliability and security of the electric bicycle by ensuring continuous control and data verification, even in the absence of certain sensors, thereby improving user safety and experience.
Implementation Method 1
The human-machine interface unit comprises at least one accelerometer (6) which is configured to determine an acceleration value which is related to an acceleration of the human-machine interface unit (4)
Implementation Method 2
The human-machine interface unit comprises at least one gyroscope (7) which is configured to determine an angular rate value which is related to an angular rate of the human-machine interface unit (4)
Data Source
Figure 1
AI summary
The present invention relates to an electric bicycle (1) and to an operating method for a control unit (5) of a human-machine interface unit (4) of such an electric bicycle (1). The present invention is based on the general concept that, any operational parameter from the different ECUs of the electric bicycle (1) together with potential acceleration data and potential gyroscope data provided by a system unit of the electric bicycle (1) to control of the electric bicycle (1).