Abstract:In this study, a gait phase classification method based on SVM multiclass classification is introduced, with a focus on the precise identification of the stance and swing phases, which are further subdivided into seven phases. Data from individual IMU sensors, such as Shank Acceleration X, Y, Z, Shank Gyro X, and Knee Angles, are used as features in this classification model. The suggested technique successfully classifies the various gait phases with a significant accuracy of about 90.3%. Gait phase classification is crucial, especially in the domains of exoskeletons and prosthetics, where accurate identification of gait phases enables seamless integration with assistive equipment, improving mobility, stability, and energy economy. This study extends the study of gait and offers an effective method for correctly identifying gait phases from Shank IMU sensor data, with potential applications in biomechanical research, exoskeletons, rehabilitation, and prosthetics.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is **how to accurately classify each stage in the gait cycle, especially for the application of assistive devices such as prostheses and exoskeletons**. Specifically, the paper aims to accurately identify seven sub - stages (Midswing, Terminal Swing, Loading Response, Midstance, Terminal Stance, Preswing and Initial Swing) in the gait cycle based on calf IMU sensor data by using the Support Vector Machine (SVM) multi - class classifier. The solution of these problems is crucial for improving the performance of prosthetic and exoskeleton devices, which can enhance the user's mobility, stability and energy efficiency.
### Specific background and importance of the problem
1. **Importance of gait analysis**:
- Gait refers to the way humans walk, involving the coordinated movement of multiple joints and muscles. Different stages of the gait cycle correspond to specific stages in the walking pattern. Accurately understanding these stages is very important for analyzing and interpreting biomechanical patterns related to prostheses.
- For amputees, gait - phase classification can help researchers and clinicians evaluate the performance of prostheses, optimize their design and control mechanisms, and achieve more natural and efficient walking.
2. **Limitations of existing methods**:
- Previous gait - phase classification methods may rely on complex sensor systems or algorithms, and have problems such as insufficient accuracy or difficulty in real - time application. Therefore, it is of great significance to develop an efficient and accurate classification method.
3. **Research objectives**:
- The objective of this research is to use the SVM multi - class classifier, combined with calf IMU sensor data (such as Shank Acceleration X, Y, Z, Shank Gyro X and Knee Angles), to achieve high - precision classification of each stage of the gait cycle.
- The research also aims to verify the effectiveness of the proposed method and explore its potential applications in prosthetic and exoskeleton technologies.
### Formula presentation
To achieve the multi - class classification task, the research adopts the One - vs - One strategy and uses the Fine Gaussian SVM. The decision function is expressed as:
\[
f(x)=\text{sign}\left(\sum_{i = 1}^{N}\alpha_i\cdot y_i\cdot K(x_i, x)+b\right)
\]
where:
- \(\alpha_i\) is the Lagrange multiplier
- \(y_i\) is the class label
- \(K(x_i, x)\) is the Gaussian kernel function, and the calculation method is:
\[
K(x_i, x)=\exp\left(-\gamma\|x_i - x\|^2\right)
\]
- \(\gamma\) is the width parameter of the Gaussian kernel function
- \(b\) is the bias term
By adjusting the parameters \(\gamma\) and the regularization parameter \(C\), the research ensures the best performance of the model.
### Summary
The main contribution of this paper is to provide an SVM - based multi - class classification method that can accurately identify each stage in the gait cycle, thus providing strong support for the development of prosthetic and exoskeleton technologies. This method not only improves the classification accuracy, but also provides new ideas for personalized rehabilitation technologies and the design of assistive devices.