International Journal of Humanoid Robotics

Knowledge of the inertial parameters of a humanoid robot is crucial for the development of model-based controllers or realistic simulation and motion planning in dynamic situations. Inertial parameters are usually provided from CAD data and thus are inaccurate especially if the robot is modified over time. Recent results showed that the inertial parameters specific to each robot can be identified using the external ground reaction forces and moments. However, the identification accuracy intrinsically depends on the excitation properties of the recorded motion and of the system’s specific level of measurement artefact/noise. In this paper, a new method for obtaining a system’s specific pseudo-online optimal set of optimal exciting motions (OEM) is proposed. This method is based on a real-time balance controller and on an optimization process to generate OEM while handling mechanical constraints. The method was experimentally validated with an NAO humanoid robot and a laboratory grade force-plate. The method shows a reduction of 1.6 times in average of the RMS difference between measured external ground reaction forces and moments and their estimates from CAD data.

As increasingly more research efforts are geared towards creating robots that can teach and interact with children in educational contexts, it has been speculated that endowing robots with artificial empathy may facilitate learning. In this paper, we provide a background to the concept of empathy, and how it factors into learning. We then present our approach to equipping a robotic tutor with several empathic qualities, describing the technical architecture and its components, a map-reading learning scenario developed for an interactive multitouch table, as well as the pedagogical and empathic strategies devised for the robot. We also describe the results of a pilot study comparing the robotic tutor with these empathic qualities against a version of the tutor without them. The pilot study was performed with 26 school children aged 10–11 at their school. Results revealed that children in the test condition indeed rated the robot as more empathic than children in the control condition. Moreover, we explored several related measures, such as relational status and learning effect, yet no other significant differences were found. We further discuss these results and provide insights into future directions.

The aim of this paper is trying to propose an efficient method of inverse kinematics and motion generation for redundant humanoid robot arm based on the intrinsic principles of human arm motion. The intrinsic principle analysis takes into account both the skeletal kinematics and muscle strength properties. Firstly, this work analyzed the kinematic redundancy problem of a human arm. By analyzing the biological feature of a human arm, the kinematic redundancy boils down to the uncertainty of elbow position. Secondly, because the muscle’s kinematic and strength properties are critical for simulating biometric motion authentically, the muscle strength property was introduced as the criterion for configuration identification and motion generation. Three types of limb configuration, dog walking, gecko climbing, and human walking limb configuration were analyzed, and two geometrical configuration identification rules were deduced to generate biomimetic motion for humanoid robotic arms. By comparing the proposed method with other five IK methods, the proposed method significantly deduced the computing time. Finally, the configuration identification rules were used to generate motions for a 7-DoF humanoid robotic arm. The results showed that the biological rules can generate biomimetic, smooth arm motions for a redundant humanoid robotic arm.

In underactuated systems, a transition between two periodic orbits is generally characterized by slow convergence. This is due to the fact that the unactuated degree of freedom (DoF) hinders the state of the system to enter the domain of attraction of the target orbit close to the fixed point of the Poincaré Map. In this paper, we introduce an optimal control algorithm to reduce the settling time of transitions between periodic orbits of underactuated walking robots. This is achieved by utilizing the hybrid zero dynamics (HZD) framework to express the feasibility condition of the transition which can be imposed as an inequality constraint in the proposed optimal control problem. In addition, the cost function penalizes deviations from the fixed point of the target periodic orbit in the zero dynamics manifold while at the same time all dynamic and kinematic assumptions are treated as constraints. Furthermore, high magnitude torques are also penalized. The numerical results show that the proposed methodology can indeed improve the settling time compared to the transition methodology usually found in the bibliography and at the same time provide a feasible and smooth motion.

Dynamic balance has to be maintained during motions of biped systems when their feet are in contact with the ground. As a necessary condition, this indicates that the calculated zero moment point (ZMP) position should be within the specified foot support region throughout the entire motion. A critical term in the ZMP formulation is the rate of angular momentum (RAM) for each link, which should be evaluated accurately and efficiently in motion planning and simulations. In this study, we propose a recursive Lagrangian method based on Denavit–Hartenberg convention to calculate the RAM for each link and the corresponding sensitivity. This method allows the evaluation of each link’s dynamic contribution to the ZMP position. The effectiveness of the proposed approach is demonstrated by simulating bipedal motions of walking and running along with their comparison against existing approaches (direct method and global force method). The accurate RAM calculation in ZMP based on the proposed approach resulted in the improved motion trajectories and reliable ground reaction forces for high-speed bipedal motion predictions.