Vehicle Force Estimation Using Model-Based Sensor Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional power-assisted vehicles require complex and costly physical force and torque sensors, which are limited in adaptability and introduce noise and delay in force estimation, especially for vehicles pushed or pulled, necessitating a cost-effective and accurate method to quantify external force without additional sensors.
Innovation Solution
A motorized vehicle with an electric drive controlled by a calculation unit that estimates external force using a model-based approach, incorporating state estimation and sensor fusion, including a 3D acceleration sensor and yaw rate sensor, to determine road gradient and vehicle dynamics, without requiring additional force sensors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If physical force sensors or torque sensors are used to measure external force, then measurement precision is improved, but device complexity and manufacturing cost increase
Solution Approach 1:
The patent replaces physical force sensors and torque sensors with a model-based estimation system that uses existing sensors (acceleration sensor, speed sensor, motor current sensor) combined with a dynamic model of the vehicle to calculate external force. This substitution eliminates the need for additional mechanical sensing hardware while achieving accurate force measurement through computational methods.
Solution Approach 2:
The patent creates a virtual model (digital twin) of the vehicle's dynamic behavior that replicates the physical system's response to various forces. By measuring the deviation between the model's predicted behavior and actual behavior, the system estimates external force without direct physical sensing, effectively copying the system's dynamics in software rather than hardware.
2Measurement precision
If physical force sensors are installed in the drawbar or handlebar, then external force measurement is improved, but adaptability decreases when pushing or pulling mode changes
Solution Approach 1:
The patent creates a universal force estimation system that works for both pulling and pushing modes without requiring mode-specific sensors. The dynamic model automatically adapts to different operating modes by interpreting the same sensor data (acceleration, speed, motor current) in the context of the vehicle's overall dynamics, making the system versatile across different usage scenarios.
Solution Approach 2:
Instead of measuring force directly at the point of application (drawbar or handlebar), the patent inverts the approach by measuring the vehicle's response (acceleration, speed changes, motor current) and working backwards to estimate the external force. This indirect measurement approach eliminates the need for mode-specific sensor locations.
3Device complexity
If conventional force estimation methods are used, then device complexity is reduced, but noise and delay in force estimation increase
Solution Approach 1:
The patent implements a feedback mechanism where the dynamic model continuously compares its predicted vehicle behavior with actual sensor measurements. The difference (residual) between predicted and actual behavior provides feedback that is used to refine the external force estimation in real-time, reducing noise and delay through continuous correction rather than simple open-loop calculation.
Solution Approach 2:
The patent performs preliminary calculations by pre-establishing the dynamic model and its parameters offline. During operation, only the final force estimation calculation is needed, which reduces computational delay. The model's structure and relationships are prepared in advance, allowing rapid real-time estimation without complex on-the-fly computations.
Data Source
Figure 1
Figure 2
AI summary
The invention describes a force estimation device for a motorized vehicle that generates an assistance torque dependent on an external force. The external force is generated directly or indirectly by a user and applied to the vehicle by pushing or pulling. To estimate this external force, a model-based state estimation is used. Based on the model parameters and system inputs (tilt angle, motor current), this first predicts the system states for the next discrete point in time. In a second step, this prediction is corrected based on the speed and/or acceleration measurements. This method for model-based force estimation enables a less noisy and more precise estimation than conventional methods that use an inverse transfer function.