A UKF-Based Predictable SVR Learning Controller for Biped Walking

The application of biped robots is always trapped by their high energy consumption. This paper makes a contribution by optimizing the joint torques to decrease the energy consumption without changing the biped gaits. In this work, a constrained quadratic programming (QP) problem for energy optimization is formulated. A neurodynamics-based solver is presented to solve the QP problem. Differing from the existing literatures, the proposed neurodynamics-based energy optimization (NEO) strategy minimizes the energy consumption and guarantees the following three important constraints simultaneously: (i) the force-moment equilibrium equation of biped robots, (ii) frictions applied by each leg on the ground to hold the biped robot without slippage and tipping over, and (iii) physical limits of the motors. Simulations demonstrate that the proposed strategy is effective for energyefficient biped walking.

show abstract

“…The performances of the proposed NEO are compared with those of the SVM control methods [40,41] and NN control methods.…”

Section: Biped Walking Via Different Controllersmentioning

confidence: 99%

“…Biped robots receive a lot of attention from the scientists and engineers these years [1][2][3][4][5][6][7][8][9][10]. Compared with wheeled driven robots and crawler robots, biped robots have anthropomorphic adaptability to unstructured environments with relative less energy consumption.…”

Section: Introductionmentioning

confidence: 99%

Constrained Quadratic Programming and Neurodynamics-Based Solver for Energy Optimization of Biped Walking Robots

Wang

Chen

Jiang

et al. 2017

Mathematical Problems in Engineering

Self Cite

View full text Add to dashboard Cite

show abstract

“…Even if the attitude estimation is a classical problem in mobile wheeled robotics [5], [6] and unmanned aerial vehicles (UAV) [7], [8], [9], several researchers have extended Open Solution for Humanoid Attitude Estimation P. Pierro, C. A. Monje, N. Mansard, P. Souères, C. Balaguer such techniques to legged locomotion [10], [11], [12], [13], [14].…”

Section: Introductionmentioning

confidence: 99%

Open Solution for Humanoid Attitude Estimation through Sensory Integration and Extended Kalman Filtering

et al. 2015

View full text Add to dashboard Cite

In this paper an Extended Kalman Filter (EKF) is used in order to estimate the real state of a humanoid robot (HRP-2 robot in our case study) using the combination of the information coming from the encoders (kinematics) and from the Inertial Measurement Unit (IMU). The integration of the kinematic information into the Kalman filtering process allows a good estimation of the attitude and reduces the complexity of the problem to the use of simple kinematic transformations, even considering the existence of accelerations and mechanical flexibilities in the robot. The EKF estimator presented here is an open solution directly applicable to any humanoid robot, which is the main contribution of our approach. Experimental results are given showing the good performance of the method.Key words: Humanoid robot, Attitude estimation, Kalman filtering, Sensor integration, Kinematics simplicity Otvoreno rješenje za estimaciju položaja humanoidnog robota integracijom senzora i proširenim Kalmanovim filtrom. U ovomečlanku koristi se prošireni Kalmanov filtar (EKF) za estimaciju stanja humanoidnog robota (HRP-2 robot) kombiniranjem informacija iz enkodera (kinematika) i inercijalne mjerne jedinice (IMU). Integracijom kinematičkih informacija u proces Kalmanovog filtra ostvaruje se dobra estimacija položaja i smanjenje složenosti problema jednostavnim kinematičkim transformacijama,čak i u slučaju postojanja akceleracija te mehaničkih savijanja robota. EKF predstavljen u ovomečlanku otvoreno je rješenje primjenjivo na bilo kojem humanoidnom robotu, što je i glavni doprinos predloženog pristupa. Ekperimentalni rezultati pokazuju dobro ponašanje predložene metode.

show abstract

“…The data used was captured from actual human walk on a flat surface. A support vector regression learning controller with a UKF-based biped state prediction of the ankle and the hip to ensure ZMP stability and improved walking flexibility is carried out on a biped robot with 7 DOF on a flat surface by Wang et al [7].…”

Section: Introductionmentioning

confidence: 99%

Control of a biped robot with flexible foot on an uneven terrain

Balakrishnan

Tripathi

Sudarshan

2014

7th International Conference on Information and Automation for Sustainability

View full text Add to dashboard Cite

A technique to achieve stable walking of a biped robot on an uneven terrain is proposed in this paper. A biped structure with a flexible foot will be able to navigate uneven terrain more effectively than the same with a flat foot. The criterion used for stable walking is Zero moment point (ZMP). The geometric center of the support polygon formed by the flexible foot is taken for dynamically planning the reference trajectory at every step. A control law is derived using computed torque control and its validity is checked using simulation study. It is observed that the actual trajectory closely follows reference with the proposed control law.

show abstract

A UKF-Based Predictable SVR Learning Controller for Biped Walking

Cited by 29 publications

References 42 publications

Constrained Quadratic Programming and Neurodynamics-Based Solver for Energy Optimization of Biped Walking Robots

Constrained Quadratic Programming and Neurodynamics-Based Solver for Energy Optimization of Biped Walking Robots

Open Solution for Humanoid Attitude Estimation through Sensory Integration and Extended Kalman Filtering

Control of a biped robot with flexible foot on an uneven terrain

Contact Info

Product

Resources

About