Gait is the method of human locomotion using limbs. Recently, the analysis of human motion, specifically human gait, has received a large amount of research attention. Human gait can contain a wide variety of information that can be used in biometrics, disease diagnosis, injury rehabilitation, and load determination. In this dissertation, the development of a model-based gait analysis technique to classify external loads is presented. Specifically, the effects of external loads on gait are quantified and these effects are then used to classify whether an individual gait pattern is the result of carrying an external load or not.

First of all, the reliability of using continuous relative phase as a metric to determine loading condition is quantified by intra-class correlation coefficients (ICC) and the number of required trials is computed. The ICC(2, 1) values showed moderate reliability and 3 trials are sufficient to determine lower body kinematics under two external load conditions. Then, the work was conducted to provide the baseline separability of load carriage conditions into loaded and unloaded categories using several lower body kinematic parameters. Satisfactory classification of subjects into the correct loading condition was achieved by resorting to linear discriminant analysis (LDA). The baseline performance from 4 subjects who were not included in training data sets shows that the use of LDA provides an 88.9% correct classification over two loaded and unloaded walking conditions. Extra weights, however, can be in the form of an external load carried by an individual or excessive body weight carried by an overweight individual. The study now attempts to define the differences in lower body gait patterns caused by either external load carriage, excessive body weight, or a combination of both. It was found significant gait differences due to external load carriage and excessive body weight. Principal Component Analysis (PCA) was also used to analyze the lower body gait patterns for four loading conditions: normal weight unloaded, normal weight loaded, overweight unloaded and overweight loaded. PCA has been shown to be a powerful tool for analyzing complex gait data. In this analysis, it is shown that in order to quantify the effects of external loads for both normal weight and overweight subjects, only two principal components (PCs) are needed.

The results in this dissertation suggest that there are gait pattern changes due to external loads, and LDA could be applied successfully to classify the gait patterns with an unknown load condition. Both load carriage and excessive body weight affect lower body kinematics, but it is proved that they are not the same loading conditions. Methods in the current work also give a potential for new medical and clinical ways of investigating gait effects in osteoarthritis patients and/or obese people.