Research On Gait Pattern Of Biped Robot With Heterogeneous Legs

Biped Robot with Heterogeneous Legs is a novel robot model, which is integration of the research of biped robot and intelligent prosthesis. To some extent, it can simulate the situation of the amputee who wears prosthetics, and it can make a large number of repeated performance tests for intelligent prosthesis. It provides a scientific ideal platform for the research of intelligent prosthesis. At the same time, it can be used to study the coordination between the human body and intelligent limb, and the walking with heterogeneous legs. Research on these questions can impels the development of biped robot technology, and improve gait of biped robot effectively so that robot can walk along complicated roads like man.Following the description of the research of the prosthesis, the system components of BRHL, and the significance of BRHL study, this paper investigates the gait pattern of BRHL in detail. It consists of three parts:the gait patterns classification of BRHL, the quantitative analysis of gait pattern of BRHL and the model building of human gait.In the process of developing BRHL or intelligent prosthesis, in order to ensure the bionic leg or prosthetic leg can accurately follow the artificial limb or health limb robustly and quickly, it needs to classify gait pattern of the artificial limb or health limb. This paper collects human lower limb acceleration signals in different gait pattern by inertial sensor systems, and uses these signals as inputs of pattern classifier. After feature extraction and dimensionality reduction on these signals using wavelet decomposition and LDA techniques, this paper verifies the stability and effectiveness of this feature extraction method using two classifier models based on different principles. The experimental results show that the proposed method in this paper can classify gait pattern of healthy limb with high accuracy.In addition, in order to using feedback control technology to control BRHL or intelligent prosthesis, it needs to know some typical gait parameters when prosthesis is working. This paper collects acceleration and angular velocity signals by inertial sensors system to calculate gait parameters quantificationally. From the experimental results, the quantitative analysis algorithm proposed in this paper can effectively remove measurement noise, and get better results. When the intelligent prosthesis identifies the road condition and gets some quantitative gait information, another problem is how to adjust prosthetic gait. This paper uses the neural network to build models for the mathematical relationships between hip joint angle, knee joint angle and stride, stride frequency. This model can supply reference to modulate hip joint and knee joint angles, while the model can also be served as an online gait generator.