Hongxu Ma

Compliant Biped Walking on Uneven Terrain with Point Feet

In this paper, we aim to realize compliant biped walking on uneven terrain with point feet. A control system is designed for a 5-link planar biped walker. According to the role that each leg plays, the control system is decomposed into two parts: the swing leg control and the support leg control. The trajectory of the swing foot is generated in real-time to regulate the walking speed. By considering the reaction torque of the swing legs hip joint as disturbance, a sliding model controller is implemented at the support legs hip joint to control the torsos posture angle. In order to make sure the landing foot does not rebound after impact, the vertical contact force control is set as the internal loop of the hips height control. In simulation, the control system is tested on a virtual 5-link planar biped walker in Matlab. Finally, stable biped walking is realized on uneven terrain with roughness up to 2cm.

Study on humanoid online motion planning

A humanoid online motion planning method is proposed using description of selected variables. The proposed method changes humanoid motion planning into planning of several selected variables. And these variables are chosen to reflect the motion state of humanoid robots, including ZMP, Center of Gravity (CoG), Center of Coxa (CoC), foot position and the rotation matrices between some key frames. Then the planning flows of the variables are shown and the dynamical and physical constraints of humanoid walking are considered simultaneously. Finally, efficiency and feasibility of proposed method are validated using kinematics and dynamics simulation, and ongoing and future work is pointed out.

Design and Dynamic Walking Control of Humanoid Robot Blackmann

This paper presents the basic configuration and control method of the humanoid robot Blackmann, which is developed by the Robot Laboratory in the National University of Defense Technology, China. Blackmann is 1.55 meters tall and 63.5 Kg weight with 36 DOFs, and it was designed to be humanlike including two legs, two arms, a trunk and a head. This paper covers the mechanical design, hardware description and motion control schemes of Blackmann, and a kind of dynamical walking control scheme based on FRI point was proposed. Up to now, Blackmann has been built and tested; the embedded control system has been realized and assembled in the body of Blackmann; the basic control schemes based on FRI point and anthropomorphic walking has been realized and further study is in process

Active compliance control for a hydraulically-actuated articulated robotic leg

This paper focus on the modelling and control of a hydraulically-actuated biologically-inspired articulated robotic leg. The leg has two hydraulically-actuated degrees of freedom (DoF), the hip and knee joints, and a passive spring DoF, the ankle joint. After a brief description of the prototype leg, the paper developed a nonlinear model of the leg, including its kinematics and dynamics. Subsequently, an actively compliant control architecture based on outer virtual compliant leg model and inner torque controller was designed to achieve a spring-damper like touchdown dynamics behavior. Lastly, both a Pushdown experiment and a Dropdown experiment were carried to verify the designed controller. The result confirmed that a hydraulically-actuated robot articulated leg can successfully emulate virtual passive components through proper mechanical design and actively compliant control under dynamic situations.

Joint torque and velocity optimization for a redundant leg of quadruped robot

Hydraulic actuated quadruped robot similar to BigDog has two primary performance requirements, load capacity and walking speed, so that it is necessary to balance joint torque and joint velocity when designing the dimension of single leg and controlling its motion. On the one hand, because there are three joints per leg on sagittal plane, it is necessary to firstly optimize the distribution of torque and angular velocity of every joint on the basis of their different requirements. On the other hand, because the performance of hydraulic actuator is limited, it is significant to keep the joint torque and angular velocity in actuator physical limitations. Therefore, it is essential to balance the joint torque and angular velocity which have negative correlation under the condition of constant power of the hydraulic actuator. The main purpose of this article is to optimize the distribution of joint torques and velocity of a redundant single leg with joint physical limitations. Firstly, a modified optimization criterion combining joint torques with angular velocity that takes both support phase and flight phase into account is proposed, and then the modified optimization criterion is converted into a normal quadratic programming problem. A kind of recurrent neural network is used to solve the quadratic program problem. This method avoids tremendous matrix inversion and fits for time-varying system. The achieved optimized distribution of joint torques and velocity is useful for aiding mechanical design and the following motion control. Simulation results presented in this article confirm the efficiency of this optimization algorithm.

Velocity control for trotting of a NUDT quadruped robot

In this paper, we propose a control strategy for the quadruped robot in the trotting gait. The quadruped robot is simplified as a planar seven-link closed kinematic chain and a LIP (linear inverted pendulum). The quadruped robot can achieve steady locomotion by velocity control. This paper designs a method of virtual sliding mode controller for estimation of velocity and applys to the method of velocity control. Simulations and experimental results demonstrate the method of control and estimation of velocity as a quadruped robot successfully trotting.

Single leg operational space control of quadruped robot based on reinforcement learning

Due to the strongly environmental adaptation, quadruped robot becomes a research hot spot. The single leg control lays foundation for the control of quadruped robot. The operational space control which is used in the quadruped robot single leg control algorithms usually strongly depends on the accuracy of the model. This paper applies RBF (radial basis function) neural network adaptive control on the single leg, and a kind of control parameters adjustment method which is based on reinforcement learning is proposed. The result shows the algorithm effectively improves the control accuracy and convergence speed under the high-dynamic condition.

Height control for a two-mass hopping robot based on stable limit cycle

Differing from the commonly used spring loaded inverted pendulum model, this paper makes use of a two-mass spring model considering impact between the foot and ground which is closer to the real hopping robot. The height of upper mass which includes the upper leg and body is the main control objective. Then we develop a new kind of control algorithm acting on two levels: The upper level aims to achieve the desired velocity of the upper mass based on a stable limit cycle, where three different controllers are used to regulate the limit cycle; the target of the lower level is to drive the system to converge to the desired state and control the contact force between the foot and ground within an appropriate range based on the inner force control at the same time. Simulation results presented in this paper confirm the efficiency of this control algorithm.

Robust walking control of a planar spring mass biped robot

The variable spring-loaded inverted pendulum (V-SLIP) model captures characteristic properties of the hip or COM motion in human locomotion. A control strategy consists of a leg stiffness controller and a foot placement controller is proposed for a biped walker. Some restrictions are considered in the control strategy, e.g. the friction and the ability of the actuator. A novel trajectory function is designed for stance phase control. The function not only can approximate the nominal trajectory with error in the order of sub-millimeter, but also can preserve the restriction on vertical velocity in spite of the horizontal velocity. To validate the control strategy and the trajectory function, simulations are implemented on a virtual ideal biped walker. The walker starts walking with a low velocity, by taking a few steps it comes to the desired walking cycle. With the proposed control strategy the walker is able to recover from a disturbance up to 20 N*m.

A BASIC VARIABLES SET BASED SCHEME OF ONLINE MOTION PLANNING FOR HUMANOID ROBOTS

An online motion planning scheme for humanoid robots is proposed based on basic variables. The proposed scheme transforms the problem of humanoid motion planning into the planning of a basic variables set. The chosen basic variables can reflect the motion state of humanoid robots, including Zero Moment Point (ZMP), Center of Gravity (COG), Center of Coxa (CoC), foot position and the rotation matrices between some key frames. Then the planning flow of basic variables are shown and the dynamical and physical constraints of humanoid motion are taken into account. Finally, the efficiency and feasibility of proposed scheme are validated by computer simulation.