Ko Yamamoto

Maximal output admissible set for trajectory tracking control of biped robots and its application to falling avoidance control

Humanoid robots have been considered as a universal machine which can operate in place of human. This kind of universal machine requires human-like biped walking capability. In particular, it is important to avoid falling by appropriately switching behaviors even if there are unknown disturbances. The authors proposed the maximal output admissible (MOA) set for the center of gravity (COG) regulator in the upright position. Based on the MOA set, we can switch feedback gains with the Zero Moment Point (ZMP) constraint satisfied. In this paper, the author extends MOA set framework to trajectory tracking controller. This extension makes it possible to switch controllers: regulator in the upright position and tracking controller of a stepping motion in order to avoid falling. The effectiveness of the proposed method is verified with a simulation.

Continuum model of crossing pedestrian flows and swarm control based on temporal/spatial frequency

In the densely-populated urban areas, pedestrian flows often cross each other and congestion occurs. It may cause discomfort feeling or pedestrian accidents. In order to reduce the congestion or the risk of accidents, it is required to control swarm flows of pedestrian. This paper proposes an implicit control method of the crossing pedestrian flows. Pedestrian flow is modeled with the continuum fluid model and its congestion degree is calculated as the fluid density. From a simulation of the crossing flows with the continuum model, it is verified that diagonal stripe pattern of the congestion degree emerges. Moreover, the authors propose an implicit control method to improve average flow velocity by moving guides. Focusing on periodic phenomenon of the crossing flows, we investigate the relationship between its temporal and spatial frequency and a periodic motion of guides. From this relationship, a control method based on the temporal and spatial frequency is proposed.

Control strategy switching for humanoid robots based on maximal output admissible set

Human-like bipedal walking is a goal of humanoid robotics. It is especially important to provide a robust falling prevention capability by imitating the human ability to switch between control strategies in response to disturbances, e.g., standing balancing and stepping motion. However, the motion control of a humanoid robot is challenging because the contact forces are constrained. This paper proposes a novel framework for control strategy switching based on the maximal output admissible (MOA) set, which is a set of initial states that satisfy the constraints. This makes it possible to determine whether the robot might fall down due to a constraint violation. The MOA set is extended to a trajectory tracking controller with a time-variant reference and constraint. In this extension, the motion of the vertical center of gravity is also considered, which has often been neglected in previous studies. Utilizing the MOA set, an example is shown of the falling prevention control by switching the standing balance control and trajectory tracking control to a stepping and hopping motion. Moreover, a method is presented for applying the MOA set framework to a position-controlled humanoid robot. The validity of the MOA set framework is verified based on simulations and experimental results. A novel framework for control strategy switching based on the maximal output admissible set (MOA) set is proposed.The formulation and computation procedure for the MOA set are presented for trajectory tracking control.The MOA set was applied to the falling prevention control. By switching between the regulator and trajectory tracking controller based on the MOA set, the robot can avoid falling with the COP constraint satisfied.An experimental computation method for the MOA set via identification of the macroscopic feedback gain is proposed. The validity of this framework was verified by the results of experiments.

Control of swarm behavior in crossing pedestrians based on temporal/spatial frequencies

In the densely-populated urban areas, pedestrian flows often cross each other and congestion is caused. The congestion makes us feel uncomfortable and sometimes leads to pedestrian accidents. To reduce the congestion or the risk of accidents, it is required to control the swarm behavior of pedestrian flows. This paper proposes modeling and controlling method of the crossing pedestrian flows. In the social/urban engineering, it is well known that the swarm behavior with a diagonal stripe pattern emerges in the crossing area of the flows. This is a self-organized phenomenon caused by the local collision avoidance effect of the pedestrians. To control the macroscopic behavior of the flows, we utilize this self-organized phenomenon. Firstly, we propose the continuum model of the crossing pedestrian flows. In the continuum model, the dynamic change of the congestion in the diagonal stripe pattern is simulated as the density. Secondly, the novel control method to improve average flow velocity is proposed based on the model. The proposed method utilizes the dynamic interaction between the diagonal stripe pattern and guides, who are moving in the flows. The authors derive the control algorithm through an analysis on the temporal and spatial frequencies of the crossing flows. The validity is verified with simulations using the continuum model. Moreover, we apply the proposed method to the particle model, assuming the actual pedestrians.

Maximal Output Admissible set for limit cycle controller of humanoid robot

This paper addresses a novel computation method of the Maximal Output Admissible (MOA) set for a limit cycle controller and its application to motion transition. By approximately calculating the MOA set via sample point cloud, we can obtain an analytic form of the MOA set even on a nonlinear system. Formulation provided in this paper can be applicable to various types of controllers. Using the MOA set, we demonstrate a motion transition from a standing posture to steady walking. Switching two types of controllers based on the MOA set realizes motion transition with the COP constraint satisfied. The validity of proposed method is verified with a simulation.

Falling prevention of humanoid robots by switching standing balance and hopping motion based on MOA set

This paper addresses falling prevention by a humanoid robot which adaptively switches a standing balance controller and a hopping motion controller. In the previous research, the author proposed a switching framework between a standing balance controller and stepping motion controller based on the Maximal Output Admissible (MOA) set. Different from stepping or walking motion, a hopping motion requires control of the COG in the vertical direction. In this paper, the MOA set is extended so as to deal with the vertical COG dynamics. The effectiveness is validated with a simulation of falling prevention.

Identification of macroscopic feedback gain in a position-controlled humanoid robot and its application to falling detection

Motion control of a humanoid robot is challenging problem because its dynamics is complicated. To make it easier to design a controller, a macroscopic dynamics focusing on the relationship between the center of gravity (COG) and the center of pressure (COP) is often used. Based on the COG-COP model, it is possible to control the COG by using the COP as the control input. Moreover, it is possible to detect falling based on the Maximal Output Admissible (MOA) set as the author presented in the previous paper. However, generating appropriate control input (COP) requires sophisticated joint actuator which can realize torque control and backdrivability. Research on this type of actuators is still on-going in the robotics field, and most existing humanoid robots are position-controlled. In this paper, the author identifies a macroscopic feedback gain in a position-controlled humanoid by measuring a disturbance response. From the obtained feedback gain, it is possible to compute the MOA set and apply it to the falling prevention control. The effectiveness of the proposed method is verified with experiments.

Robust walking by resolved viscoelasticity control explicitly considering structure-variability of a humanoid

This paper discusses the resolved viscoelasticity control (RVC) method that explicitly considers the structure-variability for a humanoid. In a previous report, the author proposed resolving the virtual viscoelasticity at the center of gravity into the joint viscoelasticity considering redundant degrees of freedom, and named this method as RVC. However, the author considered only the single support phase; therefore, the humanoid could be regarded as an open kinematic chain and the RVC was implemented easily. In this paper, the author extends the previous work on the RVC by considering structure-variability — the method now considers an open kinematic chain in the single support phase and a closed kinematic chain in the double support phase. This extension helps realize stable and robust walking motion on uneven terrains. The proposed method is validated using forward dynamics simulations.

Humanoid motion analysis and control based on COG viscoelasticity

Abstract This paper proposes a concept of center of gravity (COG) viscoelasticity to associate joint viscoelasticity with the inverted pendulum model of humanoid dynamics. Although COG viscoelasticity is based on the well-known kinematic relationship between joint stiffness and end-effector stiffness, it provides practical advantages for both analysis and control of humanoid motions. There are two main contributions. The first is that the COG viscoelasticity allows us to analyze fall risk. In a previous study, the author proposed a fall detection method based on the maximal output admissible (MOA) set, which is computed from feedback gain of the inverted pendulum model. The COG viscoelasticity associates joint viscoelasticity with the feedback gain and allows us to compute the corresponding MOA set when an arbitrary joint viscoelasticity is given. The second contribution is that the COG viscoelasticity can be also utilized in motion control. After we design a feedback gain in the inverted pendulum model utilizing the control theory, the COG viscoelasticity can directly transform it to the joint viscoelasticity. The validity of the COG viscoelasticity is verified with whole-body dynamics simulations. Graphical Abstract

Resolved COG viscoelasticity control of a humanoid

This paper proposes the concept of the center of gravity (COG) viscoelasticity to associate joint viscoelasticity with the COG-zero moment point (ZMP) model of humanoid dynamics. Although COG viscoelasticity is based on the well-known kinematic relationship between joint stiffness and end-effector stiffness, it provides practical advantages for humanoid motion control. Once the feedback gain in the COG-ZMP model is designed using the control theory, COG viscoelasticity can be applied to directly transform it to joint viscoelasticity. The author names this method as resolved COG viscoelasticity control (RCVC). In particular, this paper proposes RCVC in which the null-space of the COG Jacobian is employed. The validity of the RCVC is verified by simulating whole-body dynamics.