Recent years have witnessed a growing interest in lower limb exoskeletons (LLEs) designed to restore walking capabilities for individuals with motor impairments. Currently, the research community faces the challenge of enhancing the translational impact of this technology by proposing solutions to move from laboratory and clinical settings to daily life scenarios. In this context, the topic of environment-adaptive gait planning (EAGP) is of particular interest and importance. EAGP refers to the ability of LLEs to assist in walking on various terrains (e.g., stairs and slopes) and under different conditions (e.g., the presence of obstacles and uneven ground) by adapting their behavior according to the context. However, the proliferation of publications has led to a general lack of convergence within the scientific community toward a common definition of this research topic. Therefore, this article proposes a review of EAGP for LLEs. The objectives are to: 1) provide a clear definition of EAGP, highlighting its main components and characteristics; 2) propose a taxonomy for classifying the most frequently used techniques in the state of the art; and 3) conduct a comprehensive survey of the literature, situating each article within the proposed taxonomy. In addition, we identify the most urgent future challenges toward the development of more autonomous LLEs.

Covert visuospatial attention (CVSA) enables users to direct focus without eye movements, offering a promising non-invasive control modality for brain-machine interface (BMI) systems. Traditional CVSA-based BMIs rely on decoding α-band electroencephalography (EEG) activity from whole-trial data, often overlooking the temporal dynamics of attentional processing. In this study, we propose a novel control framework that explicitly separates the CVSA task into two subcomponents: shift attention and sustained attention. EEG signals were recorded from nine participants performing a CVSA task in a real-world environment. Through spectral feature analysis and temporal segmentation, we identify distinct patterns associated with each subcomponent and train dedicated classifiers accordingly. Our pseudo-online analysis suggests that combining the outputs of two task-specific classifiers significantly improves both trial-level classification accuracy (by an average of +23.74%) and area under the curve (AUC) (by +4.49%) compared to the conventional approach. These findings suggest that modelling the temporal structure of CVSA can enhance decoding performance and offer a more robust foundation for attention-based BMI.

Safe and effective powered wheelchair (PW) use requires training, which is typically conducted in clinical settings with specific clinical evaluation methods, such as the Powered Mobility Program (PMP). However, these methods lack objective performance assessment.Virtual Reality (VR) has emerged as a promising solution for safe, home-based training, also reducing the high costs for healthcare systems. Still, most existing VR-based solutions often rely on joystick or hand-tracking controls, making them inaccessible to individuals with severe upper limb impairments. To address this gap, we developed VR-PMP, a VR-based wheelchair driving simulator integrated with a hybrid Brain-Computer Interface (BCI) that combines electroencephalography (EEG) and electrooculography (EOG) signals for control. This system targets individuals with severe motor impairments but preserved cognitive function who could benefit from alternative, non-manual control methods. In this proof-of-concept study, we aimed to evaluate the feasibility of such a BCI system in a group of six adults without disabilities.Results showed that participants successfully navigated the virtual environment using the EOG-EEG BCI, achieving up to 100% accuracy in the classifier evaluation phase. These findings highlight the potential of our system as an innovative, alternative access solution for powered wheelchair mobility, paving the way for future clinical applications.

Motor imagery (MI) is a fundamental brain-machine interface (BMI) paradigm in which users learn how to modulate their brain signals to voluntarily and selectively activate specific areas of the sensorimotor cortex. The self-paced nature of this kinesthetic imagination makes MI well-suited for several human-robot interaction (HRI) scenarios. However, the performance of MI decoding is highly dependent on both the user’s expertise and the quality of the decoding algorithm. To address these challenges, we propose a method that integrates environmental data from robotic sensors to enhance the MI decoding process. Preliminary experiments indicate that this approach improves decoding accuracy and the overall performance of the brain-driven system, opening new opportunities for research on how machines can enhance the usability of BMIs systems.

Powered lower limb exoskeletons (LLEs) have emerged as wearable robots designed to augment users’ locomotion capabilities, offering mechanical support and additional power for both healthy and impaired subjects. However, current assistive exoskeletons are limited by predefined motion trajectories, hindering adaptability to unstructured environments encountered in daily life. To address this limitation, this paper proposes an environment-adaptive gait planning (EAGP) solution. The approach integrates scene understanding, pose estimation, and adaptive gait planning modules. A novel Collision-Free Foot Trajectory Generator (CFFTG) algorithm facilitates obstacle avoidance by computing collision-free foot trajectories, enhancing safety and adaptability. Through inverse kinematics, the planned trajectories are converted into angular joint trajectories for execution by low-level control. This comprehensive framework aims to enhance the adaptability and safety of LLEs, paving the way for broader real-world applications beyond clinical and research settings.

In this paper, we present a control system for the continuous teleoperation of a robotic manipulator via brain-machine interface (BMI). The proposed solution is based on shared control approach that allows the user to only focus on the operational tasks, while the low-level control details are automatically handled by the robotic intelligence. The user drives the manipulator through the imagination of limb movements (both hands vs. both feet) and a parameterized mapping function is implemented to convert the continuous BMI outputs into robot velocity commands which are sent to the shared control framework. The latter consists in: (i) a target predictor module, to infer the most probable target objects from the sequence of BMI commands; (ii) a control module based on an improved version of artificial potential fields (APF) to assist the user in reaching the target while avoiding collisions with obstacles in the environment. The system has been tested with a sample subject in a tabletop reach-to-grasp experiment with multiple target objects and obstacles achieving a success rate of 80%. The proposed system could be used in the future to help people with severe motor disabilities in performing daily life operations, such as drinking, feeding or manipulating objects.

In shared control teleoperation, the robot assists the user in accomplishing the desired task. Rather than simply executing the user’s command, the robot attempts to integrate it with information from the environment, such as obstacle and/or goal locations, and it modifies its behavior accordingly. In this article, we propose a real-time shared control teleoperation framework based on an artificial potential field approach improved by the dynamic generation of escape points around the obstacles. These escape points are virtual attractive points in the potential field that the robot can follow to overcome the obstacles more easily. The selection of which escape point to follow is done in real time by solving a soft-constrained problem optimizing the reaching of the most probable goal, estimated from the user’s action. Our proposal has been extensively compared with two state-of-the-art approaches in a static cluttered environment and a dynamic setup with randomly moving objects. Experimental results showed the efficacy of our method in terms of quantitative and qualitative metrics. For example, it significantly decreases the time to complete the tasks and the user’s intervention, and it helps reduce the failure rate. Moreover, we received positive feedback from the users that tested our proposal. Finally, the proposed framework is compatible with both mobile and manipulator robots.

The increasing interest of the research community in the intertwined fields of brain-machine interface (BMI) and robotics has led to the development of a variety of brain-actuated devices, ranging from powered wheelchairs and telepresence robots to wearable exoskeletons. Nevertheless, in most cases, the interaction between the two systems is still rudimentary, allowing only an unidirectional simple communication from the BMI to the robot that acts as a mere passive end-effector. This limitation could be due to the lack of a common research framework, facilitating the integration of these two technologies. In this scenario, we proposed ROS-Neuro to overcome the aforementioned limitations by providing a common middleware between BMI and robotics. In this work, we present a working example of the potentialities of ROS-Neuro by describing a full closed-loop implementation of a BMI based on motor imagination. The paper shows the general structure of a closed-loop BMI in ROS-Neuro and describes the specific implementation of the packages related to the proposed motor imagery BMI, already available online with source codes, tutorials and documentations. Furthermore, we show two practical case scenarios where the implemented BMI is used to control a computer game or a telepresence robot with ROS-Neuro. We evaluated the performance of ROS-Neuro by ensuring comparable results with respect to a previous BMI software already validated. Results demonstrated the correct behavior of the provided packages.

Recently, the complexity analysis of brain activity has shown the possibility to provide additional information to discriminate between rest and motion in real-time. In this work, we propose a novel entropy-based machine learning method to classify between standing and walking conditions from the sole brain activity. The Shannon entropy has been used as a complexity measure of electroencephalography (EEG) signals and subject-specific features for classification have been selected by Common Spatial Patterns (CSP) filter. Exploiting these features with a linear classifier, we achieved > 85% of classification accuracy over a long period (≈ 25 min) of standing and treadmill walking on 11 healthy subjects. Moreover, we implemented the proposed approach to successfully discriminate in real-time between standing and over-ground walking on one healthy subject. We suggest that the reliable discrimination of rest against walking conditions achieved by the proposed method may be exploited to have more stable control of devices to restore locomotion, avoiding unpredictable and dangerous behaviors due to the delivery of undesired control commands.

In this paper, we propose a ROS-based system to reconstruct the motion of human upper limb based on data collected with two Myo armbands in a hybrid manner. The inertial sensors' information are fused to reconstruct shoulder and elbow kinematics. Electromyographic (EMG) signals are used to estimate wrist kinematics, to fully capture the motion of the 5-DoF (degree of freedom) user's arm. The system shows a good pose estimation accuracy compared to the XSens suit with an average RMSE of 6.61°±3.31°and a R2of 0.90±0.07.

Brain-Computer Interface systems have been widely studied and explored with adults demonstrating the possibility to achieve augmentative communication and control directly from the users' brain. Nevertheless, the study and the exploitation of the BCI in children seems to be limited. In this paper we propose and present for the first time a Brain-Robot Interface enabling children to mentally drive a robot. With this regards, we exploit the combination of a P300-based Brain-Computer Interface and a shared-autonomy approach to achieve a reliable and safe robot navigation. We tested our system in a pilot study involving five children. Our preliminary results highlight the advantages of using an accumulation framework, thanks to which the performance of the children reached the 81.67 % ± 12.7 on average in terms of accuracy. During the experiments, the shared-autonomy approach involved a low-level intelligent control on board of the robot to avoid obstacles, enabling an effective navigation also with a small number of commands.

Recent advances in the Brain-Machine Interface (BMI) and in the robotic fields allowed researchers to design a new generation of brain-actuated neuroprostheses to control a variety of assistive devices-ranging from powered wheelchairs to telepresence robots and robotic arms. However, the current interaction between BMI and robotic applications is still at its infancy. One of the reasons for the limited integration between these two technologies might be identified in the lack of a common research framework. In this scenario, ROS-Neuro may represent a solution to this challenge by providing a common middleware between BMI and robotics.In this work, we propose two packages for the ROS-Neuro middleware to acquire neural signals from different commercial acquisition devices and to store them into common data formats. We describe the usage and the main functionalities of both packages such as the provided nodes, the published and subscribed topics and the available services. Furthermore, the packages are available online together with source codes, tutorials and documentations. We evaluated the performances of the two packages to ensure the absence of delays or missing data with three commercial acquisition systems and in different operating configurations. Results demonstrated the correct behavior of the provided packages. The acquisition and recorder packages represent the first releases in the ROS-Neuro ecosystem and they aim at boosting both the shared development and the research activity in the multidisciplinary field of brain-actuated robots.

The identification of intentionally delivered commands is a challenge in Brain Computer Interfaces (BCIs) based on Sensory-Motor Rhythms (SMR). It is of fundamental importance that BCI systems controlling a robotic device (i.e., upper limb prosthesis) are capable of detecting if the user is in the so called Intentional Non-Control (INC) state (i.e., holding the prosthesis in a given position). In this work, we propose a novel approach based on the entropy of the Electroencephalogram (EEG) signals to provide a continuous identification of motion intention. Results from ten healthy subjects suggest that the proposed system can be used for reliably predicting motion in real-time at a framerate of 8 Hz with 80%±5% of accuracy. Moreover, motion intention can be detected more than 1 second before muscular activation with an average accuracy of 76%±11%.