This article evaluates the performance of individuals with cerebral palsy (CP) using an inertial human–computer interface (HCI) based on head movements and a novel facilitation algorithm which we named MouseField. In total, 17 participants from three centers specialized in CP participated in a series of tests that evaluated rapidity and accuracy in pointing tasks using a head mouse named ENLAZA, with and without MouseField. Cluster analysis was used to classify the participants into three categories, C1–C3 ordered by throughput. MouseField improved rapidity in C2 and C3 and accuracy in C3. Although the accuracy of the least impaired (C1, MACS IV–V) decreased, MouseField improved task performance of users with severe motor disorders (MACS V), with decrements of 33.3 and 65.5% in the task execution time. The results show that severely impaired individuals who cannot use traditional input devices could benefit from using MouseField and ENLAZA for a facilitated navigation that minimizes targeting errors.

The computer is considered a very useful tool for physical and cognitive rehabilitation. However, people with disabilities have limitations to control properly conventional user interfaces because of their motor diseases. This paper is in the framework of a project to create an alternative computer interface, especially addressed to children with Cerebral Palsy (CP). The interface is a head mouse based on inertial technology. Although the inertial interface succeeds as mouse pointer for healthy users, children with CP have involuntary movements, such as spasms or tremor, which limit the control. In this paper, the inertial interface is used to analyze the pathological patterns of three children with CP (hypertonic-athetoid, hypertonic-dystonic and hypotonic-athetoid cases) in terms of kinematic parameters and spectral analysis of the pathological motion. Finally, two filtering techniques are designed and evaluated to reduce the effects of the involuntary motion and improve the accessibility of the interface as mouse pointer.

The application of a repetitive controller with zero phase error tracking compensation (ZPETC) to the speed control problem of travelling wave ultrasonic motors is investigated. The proposed control strategy is aimed at reducing the typical speed ripples of ultrasonic motors as well as compensating the load and temperature changes. Repetitive controllers are capable of improving the response of dynamical systems under periodic disturbances by learning from the stored error valves of preceding periods. Since it is desirable for the stabilizing compensator of repetitive controllers to be zero phase and perfect error tracker, the ZPET controller is a natural choice when the servo's closed-loop transfer function has non-minimum zeros. The performance of the proposed controller is compared against that of a PI-repetitive controller and its effectiveness experimentally demonstrated.

Application of robotics to agriculture involves specific design of manipulators to account for the need of light structures working at high speeds and acceleration, while preserving the environment against damage. All these requirements, mainly the speed and acceleration, define the choice of manipulator's structure and optimize the mechanical design for the manipulator dynamics. This paper presents the specific design of a manipulator for harvesting fresh fruit (oranges). First, we discuss the optimum manipulator structure taking into account all the necessary requirements. We then use, in order to improve the resulting dynamics, the nonlinear performance index defined in a our previous work, to reflect the nonlinear effects on the actuators due to internal parameters of the structure, and extend it to take into account the changing payload for an improved design.