Sports robots have become a popular research topic in recent years. For table-tennis robots, ball tracking and trajectory prediction are the most important technologies. Several methods were developed in previous research efforts, and they can be divided into two categories: physical models and machine learning. The former use algorithms that consider gravity, air resistance, the Magnus effect, and elastic collision. However, estimating these external forces require high sampling frequencies that can only be achieved with high-efficiency imaging equipment. This study thus employed machine learning to learn the flight trajectories of ping-pong balls, which consist of two parabolic trajectories: one beginning at the serving point and ending at the landing point on the table, and the other beginning at the landing point and ending at the striking point of the robot. We established two artificial neural networks to learn these two trajectories. We conducted a simulation experiment using 200 real-world trajectories as training data. The mean errors of the proposed dual-network method and a single-network model were 39.6 mm and 42.9 mm, respectively. The results indicate that the prediction performance of the proposed dual-network method is better than that of the single-network approach. We also used the physical model to generate 330 trajectories for training and the simulation test results show that the trained model achieved a success rate of 97% out of 30 attempts, which was higher than the success rate of 70% obtained by the physical model. A physical experiment presented a mean error and standard deviation of 36.6 mm and 18.8 mm, respectively. The results also show that even without the time stamps, the proposed method maintains its prediction performance with the additional advantages of 15% fewer parameters in the overall network and 54% shorter training time.

Compliance control is important for the realization of disturbance absorption in biped robots. However, under a sustained disturbance, compliance control causes the robot’s balance to deteriorate because of its floating base nature. Humans address this problem by resisting external disturbance. When pushed, a human will reconcile their posture with the applied external force and then push back to maintain balance. Inspired by this behavior, we propose a compliance control strategy for biped robots called resistant compliance, which allows a robot to comply with the external disturbance initially and then repel the disturbance to reduce the imbalance caused by the reconciliatory motion. As a result, the robot can obtain improved environment-interaction stability and react more like a typical human, thus making both its locomotion and its interactions more stable and safer. To realize this control strategy, the virtual-mass-model (VMM) control is redesigned to unify the disturbances from unexpected external forces and an inclined floor. Then, the VMM control is combined with the linear-inverted-pendulum model to realize resistant compliant motion. Model predictive control is used to track the reference zero-moment-point trajectory, which is essential for locomotion. To validate the proposed control strategy, the method is implemented on the human-sized humanoid robot BHR-T.

Stability control for humanoid robots based on zero moment point (ZMP) control and impedance control are widespread. However, uncertain changes in the center of mass (CoM) height for ZMP control and specific regulation of the variable stiffness of impedance control have been challenging issues in previous studies. In this article, these two problems are solved by implementing fuzzy control-based regulations. First, the fuzzy ZMP controller, which regulates the feedback gains online based on the CoM height change and CoM tracking errors, is proposed. Second, we propose a fuzzy regulation law for variable stiffness, which is applied for uncertain contact situations and inspired by the pattern of human muscle stiffness. With these two methods, the ground adaptability for humanoid robots is enhanced. The proposed method is validated with experiments on a real robot platform, BHR-T.

The humanoid robot, one of the ideal unmanned systems, has the potential to perform tasks currently carried out by human workers in industry, manufacturing, service, and disaster response. In such complicated environments with unknown disturbances, compliance is important for a humanoid robot, especially a position-controlled one, to maintain balance while performing a task. This article presents a dynamic torso compliance method for position-controlled humanoid robots to adapt to unknown external disturbances while standing or walking. This method comprises two independent controllers. One controller, which is based on a linear inverted pendulum with a flywheel, is used to realize compliant behavior by controlling the torso postures according to the zero moment point. The other controller, which is based on a double inverted pendulum, is adopted to obtain the dynamic torso movements as well as maintain balance through a quadratic-programming-based controller. With the proposed method, controls for compliance and balance are decoupled, and a position-controlled robot can perform dynamic torso compliance and maintain balance while standing or walking. The performance of the proposed method is validated by simulations and experiments on a position-controlled humanoid robot (BHR-T).

This paper presents the mechanical and control system design of the latest humanoid robot platform, BHR-5, from Beijing Institute of Technology. The robot was developed as a comprehensive platform to investigate the planning and control for the fast responsive motion under unforeseen circumstances, for example, playing table-tennis. It has improvement on mechanical structure, stiffness, and reliability. An open control architecture based on concurrent multichannel communication mode of CAN bus is proposed to upgrade the real-time communication performance and the expansibility of the control system. Experiments on walking and playing table-tennis validate the effectiveness of the design.

Highly dynamic movements such as jumping are important to improve the agility and environmental adaptation of humanoid robots. This article proposes an online optimization method to realize a vertical jump with centroidal angular momentum (CAM) control and landing impact absorption for a humanoid robot. First, the robot's center of mass (CoM) trajectory is generated by nonlinear optimization. Then, a quasi-sliding mode controller is designed to ensure that the robot tracks the CoM trajectory accurately. To avoid unexpected spinning in the flight phase, a center-of-pressure-guided angular momentum controller is designed to stabilize the CAM. The modifications of CoM and CAM are realized by online optimization of dynamic components and inverse dynamics. Two quadratic programming optimizations are utilized to generate feasible contact force/torque and joint acceleration referring to uplevel CoM and CAM controllers. In addition, a viscoelastic model-based controller is designed to absorb the vibration caused by a large contact impact. A simulation and experiment of a 0.5-m high (foot lifting distance) vertical jump are achieved on a humanoid robot platform in this article (Fig. 1).