Applications of artificial intelligence in safe human-robot interactions

Stable Robotic Grasping of Multiple Objects using Deep Neural Networks

Optimal grasping points for a robotic gripper were derived, based on object and hand geometry, using deep neural networks (DNNs). The optimal grasping cost functions were derived using probability density functions for each local cost function of the normal distribution. Using the DNN, the optimum height and width were set for the robot hand to grasp objects, whose geometric and mass centre points were also considered in obtaining the optimum grasping positions for the robot fingers and the object. The proposed algorithm was tested on 10 differently shaped objects and showed improved grip performance compared to conventional methods.

A tool for the evaluation of human lower arm injury: approach, experimental validation and application to safe robotics

This paper treats the systematic injury analysis of lower arm robot–human impacts. For this purpose, a passive mechanical lower arm (PMLA) was developed that mimics the human impact response and is suitable for systematic impact testing and prediction of mild contusions and lacerations. A mathematical model of the passive human lower arm is adopted to the control of the PMLA. Its biofidelity is verified by a number of comparative impact experiments with the PMLA and a human volunteer. The respective dynamic impact responses show very good consistency and support the fact that the developed device may serve as a human substitute in safety analysis for the described conditions. The collision tests were performed with two different robots: the DLR Lightweight Robot III (LWR-III) and the EPSON PS3L industrial robot. The data acquired in the PMLA impact experiments were used to encapsulate the results in a robot independent safety curve, taking into account robot's reflected inertia, velocity and impact geometry. Safety curves define the velocity boundaries on robot motions based on the instantaneous manipulator dynamics and possible human injury due to unforeseen impacts.

G2-type SRMPC scheme for synchronous manipulation of two redundant robot arms

In this paper, to remedy the joint-angle drift phenomenon for manipulation of two redundant robot arms, a novel scheme for simultaneous repetitive motion planning and control (SRMPC) at the joint-acceleration level is proposed, which consists of two subschemes. To do so, the performance index of each SRMPC subscheme is derived and designed by employing the gradient dynamics twice, of which a convergence theorem and its proof are presented. In addition, for improving the accuracy of the motion planning and control, position error, and velocity, error feedbacks are incorporated into the forward kinematics equation and analyzed via Zhang neural-dynamics method. Then the two subschemes are simultaneously reformulated as two quadratic programs (QPs), which are finally unified into one QP problem. Furthermore, a piecewise-linear projection equation-based neural network (PLPENN) is used to solve the unified QP problem, which can handle the strictly convex QP problem in an inverse-free manner. More importantly, via such a unified QP formulation and the corresponding PLPENN solver, the synchronism of two redundant robot arms is guaranteed. Finally, two given tasks are fulfilled by 2 three-link and 2 five-link planar robot arms, respectively. Computer-simulation results validate the efficacy and accuracy of the SRMPC scheme and the corresponding PLPENN solver for synchronous manipulation of two redundant robot arms.