Shared control is a strategy used in assistive platforms to combine human and robot orders to achieve a goal. Collaborative control is a specific shared control approach, in which user's and robot's commands are merged into an emergent one in a continuous way. Robot commands tend to improve efficiency and safety. However, sometimes, assistance can be rejected by users when their commands are too altered. This provokes frustration and stress and, usually, decreases emergent efficiency. To improve acceptance, robot navigation algorithms can be adapted to mimic human behavior when possible. We propose a novel variation of the well-known dynamic window approach (DWA) that we call biomimetical DWA (BDWA). The BDWA relies on a reward function extracted from real traces from volunteers presenting different motor disabilities navigating in a hospital environment using a rollator for support. We have compared the BDWA with other reactive algorithms in terms of similarity to paths completed by people with disabilities using a robotic rollator in a rehabilitation hospital unit. The BDWA outperforms all tested algorithms in terms of likeness to human paths and success rate.

Patient condition during rehabilitation has been traditionally assessed using clinical scales. These scales typically require the patient and/or the clinician to rate a number of condition-related items to obtain a final score. This is a time-consuming task, specially if a large number of patients are involved. Furthermore, during rehabilitation, user condition is expected to change steadily in time, so assessment may require to run these scales several times to each user. To save time, much effort has been focused on developing clinical scales that require little time to be completed. This is usually achieved by measuring a reduced set of features, i.e., focusing the scales on specific features of a defined target population (Parkinson's disease, Stroke, and so on). However, these scales still require the therapist's intervention and may be tiresome for patients who have to fill them repeatedly. This paper proposes a novel approach to automatically obtain balance scales from the onboard sensors of a robotic rollator. These sensors are used to extract spatiotemporal gait parameters from patients using the rollator for support. These parameters are derived from the user forces on the rollator handles and its odometry. Resulting parameters are used to predict the Tinetti mobility clinical scale on the fly, without therapist intervention. Our approach has been validated with 19 rollator volunteers with a variety of physical and neurological disabilities at Hospital Civil (Malaga) and Fondazione Santa Lucia (Rome). Clinicians provided traditionally obtained Tinetti scores and the proposed system was used to estimate them on the fly. Results show a small root mean squared prediction error. This method can be used for any rollator user anywhere in everyday walking conditions to obtain the Tinetti scores as often as desired and, hence evaluate their progress.

Gait analysis can provide valuable information on a person’s condition and rehabilitation progress. Gait is typically captured using external equipment and/or wearable sensors. These tests are largely constrained to specific controlled environments. In addition, gait analysis often requires experts for calibration, operation and/or to place sensors on volunteers. Alternatively, mobility support devices like rollators can be equipped with onboard sensors to monitor gait parameters, while users perform their Activities of Daily Living. Gait analysis in rollators may use odometry and force sensors in the handlebars. However, force based estimation of gait parameters is less accurate than traditional methods, especially when rollators are not properly used. This paper presents an evaluation of force based gait analysis using a smart rollator on different groups of users to determine when this methodology is applicable. In a second stage, the rollator is used in combination with two lab-based gait analysis systems to assess the rollator estimation error. Our results show that: (i) there is an inverse relation between the variance in the force difference between handlebars and support on the handlebars—related to the user condition—and the estimation error; and (ii) this error is lower than 10% when the variation in the force difference is above 7 N. This lower limit was exceeded by the 95.83% of our challenged volunteers. In conclusion, rollators are useful for gait characterization as long as users really need the device for ambulation.

Mobility loss has a major impact on autonomy. Smart rollators have been proposed to enhance human abilities when conventional devices are not enough. Many human–robot interaction systems have been proposed in the last decade in this area. Comparative analysis shows that mechanical issues aside, they mainly differ in first, equipped sensors and actuators; second, input interface; third, operation modes, and fourth adaptation capabilities. This article presents a review and a tentative taxonomy of approaches during the last 6 years. In total, 92 papers have been reviewed. We have discarded works not focused on humanrobot interaction or focused only on mechanical adaptation. A critical analysis is provided after the review and classification, highlighting systems tested with their target population.

In physical Human–Robot Interaction (pHRI), forces exerted by humans need to be estimated to accommodate robot commands to human constraints, preferences, and needs. This paper presents a method for the estimation of the interaction forces between a human and a robot using a gripper with proprioceptive sensing. Specifically, we measure forces exerted by a human limb grabbed by an underactuated gripper in a frontal plane using only the gripper’s own sensors. This is achieved via a regression method, trained with experimental data from the values of the phalanx angles and actuator signals. The proposed method is intended for adaptive shared control in limb manipulation. Although adding force sensors provides better performance, the results obtained are accurate enough for this application. This approach requires no additional hardware: it relies uniquely on the gripper motor feedback—current, position and torque—and joint angles. Also, it is computationally cheap, so processing times are low enough to allow continuous human-adapted pHRI for shared control.