Biped Robot Locomotion Research Articles

Comparative Analysis of Reinforcement Learning Algorithms for Bipedal Robot Locomotion

Comparative Analysis of Reinforcement Learning Algorithms for Bipedal Robot Locomotion

A Deep Learning Approach for Biped Robot Locomotion Interface Using a Single Inertial Sensor.

In this study, we introduce a novel framework that combines human motion parameterization from a single inertial sensor, motion synthesis from these parameters, and biped robot motion control using the synthesized motion. This framework applies advanced deep learning methods to data obtained from an IMU attached to a human subject's pelvis. This minimalistic sensor setup simplifies the data collection process, overcoming price and complexity challenges related to multi-sensor systems. We employed a Bi-LSTM encoder to estimate key human motion parameters: walking velocity and gait phase from the IMU sensor. This step is followed by a feedforward motion generator-decoder network that accurately produces lower limb joint angles and displacement corresponding to these parameters. Additionally, our method also introduces a Fourier series-based approach to generate these key motion parameters solely from user commands, specifically walking speed and gait period. Hence, the decoder can receive inputs either from the encoder or directly from the Fourier series parameter generator. The output of the decoder network is then utilized as a reference motion for the walking control of a biped robot, employing a constraint-consistent inverse dynamics control algorithm. This framework facilitates biped robot motion planning based on data from either a single inertial sensor or two user commands. The proposed method was validated through robot simulations in the MuJoco physics engine environment. The motion controller achieved an error of ≤5° in tracking the joint angles demonstrating the effectiveness of the proposed framework. This was accomplished using minimal sensor data or few user commands, marking a promising foundation for robotic control and human-robot interaction.

A Novel Low-Cost ZMP Estimation Method for Humanoid Gait using Inertial Measurement Devices: Concept and Experiments

Estimation and control of zero-moment point (ZMP) is a widely used concept for planning the locomotion of bipedal robots and is commonly measured using integrated joint angle encoders and foot force sensors. Contemporary methods for ZMP measurement involve built-in contact sensors such as joint encoders or instrumented foot force sensors. This paper presents a novel approach for computing ZMP for a humanoid robot using inertial sensor-based wireless foot sensor modules (WFSMs). The developed WFSMs, strapped at different limb segments of a bipedal robot, measure lower limb joint angles in real time. The joint angle trajectories, further transformed into Cartesian position coordinates, are used for estimating the ZMP positions of humanoid robots using the planar biped model. The whole framework is presented through experimental studies for different real-life walking scenarios. Since the modules work based on the limb motion and inclination, any ground unevenness would be automatically reflected in the module output. Hence, this measurement process can be a convenient method for applications requiring humanoid control on uneven surfaces/outdoor terrains. To compare the performance of the proposed model, ZMP is simultaneously measured from inbuilt foot force sensors and joint encoders of the robot. Statistical tests exhibit a high linear correlation between the proposed method with integrated encoders and foot force sensors (Pearson’s coefficient, [Formula: see text]). Results indicate that ZMP estimated by WFSM is a viable method to monitor the dynamic gait balance of a humanoid robot and has potential application in outdoor and uneven terrains.

Risk-Averse Control via CVaR Barrier Functions: Application to Bipedal Robot Locomotion

Enforcing safety in the presence of stochastic uncertainty is a challenging problem. Traditionally, researchers have proposed safety in the statistical mean as a safety measure for systems subject to stochastic uncertainty. However, ensuring safety in the statistical mean is only reasonable if system's safe behavior in the large number of runs is of interest, which precludes the use of mean safety in practical scenarios. In this letter, we propose a risk sensitive notion of safety called conditional-value-at-risk (CVaR) safety. We introduce CVaR barrier functions as a tool to enforce CVaR-safety and propose conditions for their Boolean compositions. Given a legacy controller, we show that we can design a minimally interfering CVaR-safe controller via solving difference convex programs (DCPs). We elucidate the proposed method by applying it to a bipedal robot locomotion case study.

Deep ensemble learning approach for lower extremity activities recognition using wearable sensors

AbstractHuman walking is a very challenging task and always requires rigorous practice. It is a learning process that involves the complex coordination of the brain and lower limbs. The bipedal robots that mimic the human morphological structure to produce human similar walking, are not capable of producing an efficient walk. Due to walking challenges and structural differences, a robot cannot walk like a human being. In this research, to achieve the aforementioned objective to produce a human similar walk, human lower extremity activities are considered to understand walking behaviour. The experiment involves different walking styles on different terrains. To capture the learning process of bipedal robot locomotion, a deep learning‐based ensemble classifier is introduced for human lower activities recognition. To understand the learning process seven different walking activities are considered for analysis purposes. An Inertial measurement unit (IMU) is used as a wearable device due to its small form factor and unobtrusive nature to capture the walking movement of different lower limbs joints. Three public datasets viz. mHealth, OU‐ISIR similar action and HAPT inertial sensor data sets are considered for this study. To classify the activities, 2 different deep learning models namely convolutional neural network (CNN) and long short‐term memory (LSTM) are used. To generalize the results, an ensemble of different classifiers is implemented. The Classifier has reported accuracy of 99.25%, 88.48% and 97.44%, respectively, on the aforementioned data sets. This work can be utilized for elderly subjects' postural stability, rehabilitation of patients post‐stroke and trauma, generation of robot walk trajectories in cluttered environment and reconstruction of impaired walking.

Analysis and control of biped robot with variable stiffness ankle joints.

BACKGROUND: Biped robot locomotion is an active topic of research, and the walking stability is one of the research objectives.OBJECTIVE:This paper discusses the variable stiffness ankle joint and the walking control of a biped robot.METHODS: A design is introduced that achieves the ankle joint variable stiffness by using a pneumatic unit. The actuation system of the biped robot is based on the hybrid use of electric and pneumatic. The locomotion control architecture has been proposed to exploit natural leg dynamics in order to improve the biped robot walking stability. We also present a dynamic simulation which matches the biped robot and experiments with the real biped robot.RESULTS: The simulation and experiments result that introducing the variable stiffness ankle joint and the controller achieve a significant improvement in foot-ground impact and walking stability of the biped robot.CONCLUSION:The biped robot with variable stiffness ankle joints has a better walking performance under the control method.

A Benchmarking of DCM-Based Architectures for Position, Velocity and Torque-Controlled Humanoid Robots

This paper contributes toward the benchmarking of control architectures for bipedal robot locomotion. It considers architectures that are based on the Divergent Component of Motion (DCM) and composed of three main layers: trajectory optimization, simplified model control, and whole-body quadratic programming (QP) control layer. While the first two layers use simplified robot models, the whole-body QP control layer uses a complete robot model to produce either desired positions, velocities, or torques inputs at the joint-level. This paper then compares two implementations of the simplified model control layer, which are tested with position, velocity, and torque control modes for the whole-body QP control layer. In particular, both an instantaneous and a Receding Horizon controller are presented for the simplified model control layer. We show also that one of the proposed architectures allows the humanoid robot iCub to achieve a forward walking velocity of 0.3372[Formula: see text]m/s, which is the highest walking velocity achieved by the iCub robot.

Design of the Walking Pattern Generation Software of AR-600 Anthropomorphic Platform

Biped walking implies a multitude of cyclically moving body segments and requires corresponding adjustment and coordinated movement of these segments to ensure smooth motion and balance maintenance – a challenging task considering technical limitations of currently existing robotic platforms. The article describes the implementation of the real-time dynamic walking of the AR-600 anthropomorphic robotic platform produced by JSC NPO Androidnaya Tekhnika (Magnitogorsk, Russia). The robot was controlled employing the software developed by the authors and motions were simulated in the 3DLK specialized simulation environment (JSC NPO Androidnaya Tekhnika). The authors studied various walking trajectories and confirmed the validity of this algorithm and related software for implementing biped robot locomotion.

Natural Oscillation Gait in Humanoid Biped Locomotion

Biped locomotion is complex to analyze in general. Most of the models in the literature for biped locomotion utilize hybrid dynamics. This makes the kinematics of the model simple to understand. However, it makes the model difficult to analyze in the presence of discontinuous terms in its differential equations. This article presents a continuous-time mathematical model for biped robots, modeling the discontinuities in impact with a nonlinear function. Natural oscillation is defined as a free response under damping compensation to achieve persistent oscillation. The phenomenon of natural oscillation was previously exploited in the linear dynamics of multi-segmental locomotor, and it is now extended to this continuous-time biped locomotion model with the contact force from the ground modeled as a nonlinear bump function. The idea of natural oscillation is preceded in this article by a complete framework on analysis and entrainment control. The research would ensure that these natural oscillations are responsible for forward locomotion of biped robots under certain conditions. The locomotion profile of the bipeds further validates the model that the velocity achieved through the natural oscillation is comparable to human jogging speeds when practical constraints are enforced.

On the Existence and Uniqueness of Poincaré Maps for Systems With Impulse Effects

The Poincare map is widely used to study the qualitative behavior of dynamical systems. For instance, it can be used to describe the existence of periodic solutions. The Poincare map for dynamical systems with impulse effects (SIEs) was introduced in the last decade and mainly employed to study the existence of limit cycles (periodic gaits) for the locomotion of bipedal robots. We investigate sufficient conditions for the existence and uniqueness of Poincare maps for dynamical SIEs evolving on a differentiable manifold. We apply the results to show the existence and uniqueness of Poincare maps for systems with multiple domains.

Human-Inspired Online Path Planning and Biped Walking Realization in Unknown Environment.

The focus of research in biped locomotion has moved toward real-life scenario applications, like walking on uneven terrain, passing through doors, climbing stairs and ladders. As a result, we are witnessing significant advances in the locomotion of biped robots, enabling them to move in hazardous environments while simultaneously accomplishing complex manipulation tasks. Yet, considering walking in an unknown environment, the efficiency of humanoid robots is still far from being comparable with the human. Currently, bipeds are very sensitive to external changes and they have severe constraints for adaptation of walk to conditions from such a complex environment. Promising approaches for efficient generation and realization of walking in a complex environment are based on biological solutions that have been developed for many years of evolution. This work presents one such human-inspired methodology for path planning and realization of biped walk appropriate for motion in a complex unfamiliar environment. Path planning results in calculating clothoid curves that represent well the human-like walking path. The robot walk is realized by the composition of parametric motion primitives. Such an approach enables on-line modification of planned path and walk parameters at any moment, instantly. To establish the relationship between high-level path planner and the low-level joint motion realization, we had to find a way to extract the parameters of the clothoid paths that can be linked with the parameters of the walk and consequently to motion primitive parameters. This enabled the robot to adopt its walking for avoiding the obstacles and for a smooth transition between different paths. In this paper we provide a complete framework that integrates the following components: (i) bio-inspired online path planning, (ii) path-dependent automatic calculation of high-level gait parameters (step length, walking speed, direction, and the height of the foot sole), and (iii) automatic calculation of low-level joint movements and corresponding control terms (driving motor voltage) through the adaptation of motion primitives which realize walking pattern and preserves the dynamic balance of the robot.

Evolving a Sensory–Motor Interconnection Structure for Adaptive Biped Robot Locomotion

We present an evolving neural oscillator-based bio-inspired biped robot locomotion for minimizing the constraints during the locomotion process. Sensory–motor coordination model is represented by the interconnection between motor neurons and sensory neurons. An evolutionary computation technique is applied for reconstructing the number of joints and the number of neurons in each joint depending on the environmental condition. In this system, either the number of joints, or the number of neurons, or the interconnection structure are dynamically changed depending on the conditions acquired from the sensors that equipped in the robot. Bacterial programming is inspired by the evolutionary process of bacteria, including bacterial mutation and gene transfer. This system is applied in computer simulation for realizing the optimization process and the optimized structure is applied in a small humanoid robot. In experiments, we run the robot in several different environmental conditions. Different neuron structures are resulted depending on the environmental conditions. The proposed tree structure-based optimization strategy can simplify the sensory–motor interconnection structure.

Surface Detection and Classification for Point-Footed Bipedal Robotic Locomotion in Unstructured Environments Using Depth Visual Perception

Surface Detection and Classification for Point-Footed Bipedal Robotic Locomotion in Unstructured Environments Using Depth Visual Perception

A Method for Bipedal Robot Locomotion Control along an Arbitrary Trajectory

This paper presents an algorithm for stable bipedal walking control along an arbitrary curve. The algorithm starts with a foot planner, which takes a parametrically defined desired path as an input and calculates feet positions and orientations at each step. Zero moment point (ZMP) concept is used for robot stability control. The dynamics of the robot is modeled as a running cart on a table. Given the reference ZMP trajectory, ZMP tracking servo controller based on preview control theory is used to calculate robot’s center of mass (CoM) trajectory. The preview controller is made of three terms: the integral error of ZMP, the state feedback proportional to a current state vector and the preview action which takes into account future values of the desired ZMP position. We propose robot’s state estimator based on linear Kalman filter with measured CoM acceleration and position as system inputs. Swing foot trajectories are calculated using trigonometric functions, since they are simple and can provide zero velocities at contact moments. We put some additional constraints on a system by assuming that there is no forward-backward inclination of the robot trunk and swing foot is always parallel to the ground. To avoid kinematic limits in hip and ankle roll joints we calculate a minimal angle to rotate robot trunk in frontal plane. After position and orientation of robot trunk and swing foot are found we apply inverse kinematics solution for robot legs to get robot joint angles. Finally, joint angles are sent to robot actuators. Software package based on ROS operating system was developed for AR601 robot. Verification tests were executed with robot model in Gazebo simulator. The robot successfully completed series of experiments confirming modelling results.

Research on Biped Locomotion Robot with Inertial Assist and Synchronous-drive on Self-excitation

Research on Biped Locomotion Robot with Inertial Assist and Synchronous-drive on Self-excitation

慣性アシスト付き２足歩行ロボットの研究

慣性アシスト付き２足歩行ロボットの研究

The Effect of Foot Structure on Locomotion of a Small Biped Robot

This paper is a presentation of a work that consists of considering a novel foot structure for biped robot inspired by human foot. The specific objective is to develop a foot mechanism with human-like toes for a small biped robot. The chosen architecture to present the biped includes ten degrees of freedom (DoF) on ten articulations between eleven links. Our study considers the effect of varying foot structure on a walking process of the robot in simulation by ADAMS (MSC software, USA) through gait generation method. In toe mechanism, aiming to reduce the energy consumption, the passive joint was selected as the toe joint. The center of gravity (CoG) point trajectories of the robot with varying toe is compared with each other in normal motion on flat terrain to determine the most consistent toe mechanism. The result shows that the selected foot structure enables the robot to walk stably and naturally.

Numerical Estimation of Stability Region of Self-Clocked Bipedal Robots with Inertial Disk

Bipedal robot control techniques implementing self clocking properties, as presented in Westervelt et al. (Feedback control of dynamic bipedal robot locomotion, Taylor & Francis LLC, Milton Park, 2007) and Novaes et al. (IFAC 19th world congress, 4843–4848, 2014), have desirable properties, like the ability to drive the robot step with respect to its own geometry. In this control technique, one of the possible angular coordinates of the robot is made monotonic during the step execution. The other angular coordinates will track reference functions of the former. One difficulty with this approach is to determine the region around the nominal step for which the stability is assured. In this paper, we propose an algorithm to estimate this stability region by computing numerically an interval matrix related to a Poincare return map. In conjunction with the algorithm proposed in Ahn and Chen (IEEE Trans Autom Control 52(3):510–514, 2007), we are able to determine the maximum singular value of this interval matrix and so determine a guaranteed stability region.

Development of Fuzzy Logic Controllers for Controlling Bipedal Robot Locomotion on Uneven Terrains with IMU Feedbacks

Locomotion controller is an important and essential aspect for bipedal robots. Typically, a Linear Inverted Pendulum Model (LIPM) is a mathematical approach to generate the Center of Mass (CoM) trajectory of a bipedal robot. By combining the swing foot trajectory, the omni-directional walking command is capable of generating joint angle control commands in terms of Inverse Kinematics (IK). To improve bipedal locomotion stability on uneven terrain situations, an Inertia Measurement Unit (IMU) was desired to place on the robot's chest was used to measure the body's tilt posture on uneven terrains. The robot body's tilt posture provided an indication of locomotion stability. The body's tilt posture information was further evaluated with a Fuzzy Logic Controller (FLC) to generate appropriate offset angles to be applied on the corresponding joints so that the body's tilt posture can be adjusted accordingly to meet a stable situation. Finally, a kid-size bipedal robot, named Huro Evolution JR, was used as the experiment platform. The proposed FLC can be applicable to the terrain conditions of maximum 25° slope in Double Support Phased (DSP) stand cases. With the walking cases, the FLC is capable of walking on maximum 12° slope, 1 cm stair height and the combined terrain situation well. In the future, the Center of Pressure (CoP) information will be accompanied with the IMU information to further improved the locomotion stability in a high dynamic environment.

Precise velocity control of a bipedal robot using CPG circuits in creating different behaviors

In this paper, we have extended our previous work, which created several motion phases in running and walking movements with a central pattern generator (CPG) model, to controlling the robot’s walking speed easily and precisely within a wide speed range to investigate the possibility of CPG in the domain of bipedal robotic locomotion. Our method proposed herein basically used our hybrid CPG (H-CPG) model. This model consisted of a rhythm generator using CPG, a force controller, and an attitude controller. We have also introduced a modulation mechanism using a velocity feedback loop in this force controller for the precise control of acting forces from the tip of the support leg to the floor and the walking stride length. This mechanism can realize accurate modulation of the walking velocity reflecting an error between the target velocity and the current velocity in real time. Numerical simulation results show the effectiveness of the proposed method for realizing precise control of the locomotion speed in both backward and forward stable walking.