A lack of mobility often leads to limited participation in social life. The purpose of this STREP is to conceive a system empowering lower limbs disabled people with walking abilities that let them perform their usual daily activities in the most autonomous and natural manner. New smart dry EEG bio-sensors will be applied to enable lightweight wearable EEG caps for everyday use. Novel approaches to non-invasive BCI will be experimented in order to control a purpose-designed lower limbs orthosis enabling different types of gaits. Complementary research on EMG processing will strengthen the approach. A Virtual Reality (VR) training environment will assist the patients in generating the correct brain control signals and in properly using the orthosis. The main BCI approach relies on Dynamic Recurrent Neural Network (DRNN) technology applied in a two stages process. After learning, the system will be able to match EMG signals to legs movements (Stage 2), and EEG to such EMG signals (Stage 1). The Stage 2 has already been successfully demonstrated by a project partner. The orthosis will be designed to support the weight of an adult, to address the dynamic stability of a body-exoskeleton combined system, and to enable different walking modalities. The VR training environment will comprise both a set of components for the progressive patient training under a safe and controlled medical environment, and a lightweight portable set using immersive VR solutions for self-training at home. The developed technologies will be assessed and validated with the support of a formal clinical validation procedure. This will allow to measure the strengths and weaknesses of the chosen approaches and to identify improvements required to build a future commercial system. In addition the resulting system will be progressively tested in everyday life environments and situations, ranging from simple activities at home to eventually shopping and interacting with people in the street.