To date, several alterations in the gait pattern can be treated through rehabilitative approaches and robot assisted therapy (RAT). Gait data and gait trajectories are essential in specific exoskeleton control strategies. Nevertheless, the scarcity of human gait data due to the high cost of data collection or privacy concerns can hinder the performance of controllers or models. This paper thus first creates a GANs-based (Generative Adversarial Networks) data augmentation method to generate synthetic human gait data while still retaining the dynamics of the real gait data. Then, both the real collected and the synthesized gait data are fed to our constructed two-stage attention model for gait trajectories prediction. The real human gait data are collected with the five healthy subjects recruited from an optical motion capture platform. Experimental results indicate that the created GANs-based data augmentation model can synthesize realistic-looking multi-dimensional human gait data. Also, the two-stage attention model performs better compared with the LSTM model; the attention mechanism shows a higher capacity of learning dependencies between the historical gait data to accurately predict the current values of the hip joint angles and knee joint angles in the gait trajectory. The predicted gait trajectories depending on the historical gait data can be further used for gait trajectory tracking strategies.

Cognitive impairment is one of the crucial problems elderly people face. Tracking their daily life activities and detecting early indicators of cognitive decline would be necessary for further diagnosis. Depending on the decline magnitude, monitoring may need to be done over long periods of time to detect abnormal behaviour. In the absence of training data, it would be helpful to learn the normal behaviour and daily life patterns of a (cognitively) healthy person and use them as a basis for tracking other patients. In this paper, we propose to investigate Recursive Auto-Encoders (RAE)-based transfer learning to cope with the problem of scarcity of data in the context of abnormal behaviour detection. We present a method for generating synthetic data to reflect on some behavior of people with dementia. An RAE model is trained on data of a healthy person in a source household. Then, the resulting RAE is used to detect abnormal behavior in a target house. To evaluate the proposed approach, we compare the results with the-state-of-the-art supervised methods. The results indicate that transfer learning is promising when there is lack of training data.