International Journal of Humanoid Robotics

Gait pattern performance, for its crucial significance in humanoid robot stabilization, has attracted the attention of researchers worldwide. Although simplified models highlight major features, bipedal walking has bewildered the researchers. Therefore, for a precise understanding of the bipedal model, a state-of-the-art, simplified model has been proposed in this paper which comprises a 3D-multilinked dual spring-loaded inverted pendulum (3D-MDSLIP) while acknowledging the vertical fluctuations of the center of mass (CoM). In addition, the model considers upper body movement and its effects on the stabilization of the humanoid robot. The mathematical modeling of a humanoid walking over the obstacle and slope is demonstrated to precisely understand the problem. The tuning process of the parameters and postures in a humanoid robot is complex and time-consuming. For proper walking of a robot over uneven terrains and slopes, tuning of the PID controller is achieved using converged teaching-learning based optimization (CTLBO) technique for a central pattern generator (CPG) gait, as introduced in the paper. The optimal gait angles are applied to the experimental and simulated NAO to successfully navigate the provided terrain. Thus, the experimental and simulation results jointly show that the proposed CPG-CTLBO gait learning technique is feasible for finding an optimal gait pattern for the humanoid robot within a deviation of 5%. The energy efficiency of the proposed controller is compared with the default controller of NAO based on the average electronic current in sagittal and lateral movement. Further, it is examined for the energy consumption for several slopes, and the results obtained are acceptable, showing the controller is efficient. Additionally, it has been compared with an existing technique for walking a humanoid robot on uneven terrains. The graph obtained using the proposed technique demonstrates the superiority of the proposed technique.

In this study, a novel human-in-the-loop design method using a genetic algorithm (GA) is presented to design a low-cost and easy-to-use four-bar linkage medical device for upper limb muscle rehabilitation. The four-bar linkage can generate a variety of coupler point trajectories by using different link lengths. For this medical device, patients grab the coupler point handle and rotate the arm along the designed coupler point trajectory to exercise upper limb muscles. The design procedures include three basic steps: First, for a set of link lengths, a complete coupler point trajectory is generated from four-bar linkage kinematics; second, optimization-based motion prediction is utilized to predict arm motion (joint angle profiles) subjected to hand grasping and joint angle limit constraints; third, the predicted joint angles and given hand forces are imported into an OpenSim musculoskeletal arm model to calculate the muscle forces and activations by using the OpenSim static optimization. In the GA optimization formulation, the design variables are the four-bar link lengths. The objective function is to maximize a specific muscle’s exertion for a complete arm rotation. Finally, different four-bar configurations are designed for different muscle strength exercises. The proposed human-in-the-loop design approach successfully integrates GA with linkage kinematics, arm motion prediction, and OpenSim static optimization for four-bar linkage design for upper limb muscle strength rehabilitation.

The research expects to promote “artificial intelligence (AI) $+$ Education” and cultivate high-quality art design talents with international perspectives. First, random sampling recruits 100 Art Design sophomores in a university and divides them into a test group and a control group. The Questionnaire Survey (QS) is used to obtain research data, including students’ satisfaction with the curriculum, training mode, and innovation environment. Meanwhile, some deficiencies in the traditional educational model are revealed alongside targeted suggestions. Then, a comparative analysis is performed on the design works of the two groups and the student’s evaluation of the existing and proposed models. The result proves the effectiveness of the proposed international art design talents-oriented Human–Computer Interaction (HCI)-based training model. The research results imply that students are less satisfied with the existing curriculum, training mode, and innovative environment. About 40% of students hold a neutral attitude towards the current curriculum. They hope that the school provides a more comprehensive and flexible curriculum, personalized training methods, and a relaxed learning environment conducive to creativity. Students in the test and the control groups have gained significant differences in the scores of design works. The test group has scored relatively high, and their works contain more international elements than the control group. Meanwhile, the test group’s works reflect a deeper understanding of theoretical knowledge. They give a high evaluation of the proposed talent training model. Thus, the proposed HCI-based new talent training model is effective. Therefore, the proposal is of great significance for promoting the “AI $+$ Education” and talent training models.