Closed-Loop Push Recovery for an Inexpensive Humanoid Robot

(2004). Robo-Erectus: a low-cost autonomous humanoid soccer robot. Advanced Robotics: Vol. 18, No. 7, pp. 717-720.

This paper contributes to the problem of humanoid robot gait generation in unknown environments. The intention of the proposed method is to create an autonomous humanoid robot, able to take decisions and generate the appropriate optimal gait based on the information received by the eye system. Up to now, we have created two modules: walking and going upstairs. In order to create an autonomous humanoid robot, we plan to consider other tasks like going downstairs, creeping, obstacle overcoming, etc. In this paper, we present the simulation and experimental results for real time humanoid robot gait generation realized with the “Bonten‐Maru I” humanoid robot. The results showed that the Neural Network modules generate in a very short time a stable humanoid robot motion.

Maintaining the balance is always a challenge issue for bipedal humanoid robots in dealing with various locomotive tasks to serve human society, especially when the real environment the robot worked within exhibits to be very complex. Unlike plenty of previous successful approaches on humanoid balancing or push recovery, in this research, the Dynamical Movement Primitives (DMP) is employed to model several typical bio-inspired strategies. As humanoid balancing or push recovery could be regarded as a problem of how a robot to get back to its ongoing behavior when a break happens on account of external force or uneven terrain etc., the DMP model becomes an alternative ideal choice due to its promising nature of attractor. Meanwhile, the DMP composed of a set of differential equations provides a compact parameterized representation in modelling a motion strategy, and thus leads to a strategy model that is suitable to be fulfilled with machine learning techniques. In this research, the learning process for those bio-inspired strategies modeled with DMP are completed by adopting the stochastic policy gradient reinforcement learning and imitation learning separately. Furthermore, with Gaussian Process(GP) regression, push recovery strategies are generalized taking the advantages of the invariance properties of the DMP model. As a consequence, an online adaptive push recovery control strategy is finally achieved. Experimental results on both simulated robot and a real bipedal humanoid robot PKU-HR5 demonstrate the presented approach is effective and promising.

This work is part of an ongoing work for sign language tutoring with imitation based turn-taking and interaction games with humanoid robots. The aim of the project is to motivate the children with communication problems to understand and imitate the signs implemented by the humanoid robots. In this study, we implemented selected words from Turkish Sign Language Dictionary on autonomous humanoid robots (NAO H-25 and Robovie R3). We also examined the effects of interaction games on sign language teaching between participants with different age groups and the humanoid robot.

A behavior strategy of humanoid robot for obstacle avoidance based on four infrared sensors is proposed and implemented on an autonomous humanoid robot. A mechanical structure with 26 degrees of freedom is design so that an implemented small-size humanoid robot named TWNHR-Ⅲ is able to accomplish five walking motions. Three walking experiments are presented to illustrate that the proposed biped structure lets TWNHR-Ⅲ can move forward, turn, and slip. One electronic compass and four infrared sensors are mounted on TWNHR-Ⅲ to obtain the head direction of the robot and detect obstacles, respectively. Based on the obtained information from these sensors, a decision tree method is proposed to decide one behavior from five movements: walk forward, turn right and left, and slip right and left. Two MATLAB simulations and one real experiment are presented to illustrate that the robot can avoid obstacles autonomously and go to the destination effectively.

Bipedal walking in human environments is made difficult by the unevenness of the terrain and by external disturbances. Most approaches to bipedal walking in such environments either rely upon a precise model of the surface or special hardware designed for uneven terrain. In this paper, we present an alternative approach to stabilize the walking of an inexpensive, commercially-available, position-controlled humanoid robot in difficult environments. We use electrically compliant swing foot dynamics and onboard sensors to estimate the inclination of the local surface, and use a online learning algorithm to learn an adaptive surface model. Perturbations due to external disturbances or model errors are rejected by a hierarchical push recovery controller, which modulates three biomechanically motivated push recovery controllers according to the current estimated state. We use a physically realistic simulation with an articulated robot model and reinforcement learning algorithm to train the push recovery controller, and implement the learned controller on a commercial DARwIn-OP small humanoid robot. Experimental results show that this combined approach enables the robot to walk over unknown, uneven surfaces without falling down.

This paper presents a new design for formation control on a group of autonomous humanoid robots with obstacle and collision avoidance ability that inspired by human walking behavior in a pedestrian crowd. This human walking behavior model is known as Social Force Model (SFM) and has been widely used to predict individual movement in a crowd in many variety of situation. We combine consensus algorithm with virtual structure approach for formation control and SFM for obstacle and collision avoidance. In this research, we induced some factors in SFM into consensus equation and then implemented into a group of humanoid robots. The aim of the integration is to create a group of robots that are capable of carrying out its collective tasks while still able to maintain its safety. The use of perception effect, distinct obstacle and collision avoidance function on the algorithm will differentiate algorithm with the others. The attractive feature of this new design is the various behaviors of robots to go back to their initial formation after they avoid obstacles. In the case of robot is trapped in a crowded situation (singularity condition), robot will still trying to look a new position, until it find a condition that allow him to move forward. We verify the proposed algorithm by some simulations and experiments.

Push recovery during locomotion will facilitate the deployment of humanoid robots in human-centered environments. In this paper, we present a unified framework for walking control and push recovery for humanoid robots, leveraging the arms for push recovery while dynamically walking. The key innovation is to use the environment, such as walls, to facilitate push recovery by combining Single Rigid Body model predictive control (SRB-MPC) with Hybrid Linear Inverted Pendulum (HLIP) dynamics to enable robust locomotion, push detection, and recovery by utilizing the robot's arms to brace against such walls and dynamically adjusting the desired contact forces and stepping patterns. Extensive simulation results on a humanoid robot demonstrate improved perturbation rejection and tracking performance compared to HLIP alone, with the robot able to recover from pushes up to 100 N for 0.2 s while walking at commanded speeds up to <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$0.5 ~\mathrm{m} / \mathrm{s}$</tex>. Robustness is further validated in scenarios with angled walls and multi-directional pushes.

Push recovery is an essential requirement for a humanoid robot with the objective of safely performing tasks within a real dynamic environment. In this environment, the robot is susceptible to external disturbance that in some cases is inevitable, requiring push recovery strategies to avoid possible falls, damage in humans and the environment. In this paper, a novel push recovery approach to counteract disturbance from any direction and any walking phase is developed. It presents a pattern generator with the ability to be modified according to the push recovery strategy. The result is a humanoid robot that can maintain its balance in the presence of strong disturbance taking into account its magnitude and determining the best push recovery strategy. Push recovery experiments with different disturbance directions have been performed using a 20 DOF Darwin-OP robot. The adaptability and low computational cost of the whole scheme allows is incorporation into an embedded system.

Researchers’ goals shape the questions they raise, collaborators they choose, methods they use, and outcomes of their work. This article offers a fresh vision of artificial intelligence (AI) research by suggesting a simplification to two goals: 1) emulation to understand human abilities to build systems that perform tasks as well as or better than humans and 2) application of AI methods to build widely used products and services. Researchers and developers for each goal can fruitfully work along their desired paths, but this article is intended to limit the problems that arise when assumptions from one goal are used to drive work on the other goal. For example, autonomous humanoid robots are prominent with emulation researchers, but application developers avoid them, in favor of tool-like appliances or teleoperated devices for widely used commercial products and services. This article covers four such mismatches in goals that affect AI-guided application development: 1) intelligent agent or powerful tool; 2) simulated teammate or teleoperated device; 3) autonomous system or supervisory control; and 4) humanoid robot or mechanoid appliance. This article clarifies these mismatches to facilitate the discovery of workable compromise designs that will accelerate human-centered AI applications research. A greater emphasis on human-centered AI could reduce AI’s existential threats and increase benefits for users and society, such as in business, education, healthcare, environmental preservation, and community safety.

Bipedal humanoid robots have complex dynamics and they are intrinsically unstable. Although, bipedal robots have a similar kinematic architecture to a human and they are the most appropriate type of robots to operate in human environments, developing a humanoid robot that has robust is a difficult task. This paper proposes an omnidirectional walking engine that takes into account the push recovery strategies. The walking engine has a hierarchical structure and tries to fade the complexities of the dynamic walking. In addition, it can adapt to another platform with few changes. We enhanced the Linear Inverted Pendulum Plus Flywheel Model to release the height constraint of the center of mass. This enhancement allows a more human-like motion and provides more stable walking. The proposed walking engine has been successfully tested on a real humanoid soccer robot. The experimental results show the performance of this controller to generate stable walking. The average speed of walking that we have achieved was 20cm/sec.

The human being can negotiate with external push up to certain extent reactively. Grown up persons have better push recovery capability than kids and also the professional wrestlers acquire better push recovery capability than normal human being. The acquired push recovery capability, therefore, is based on learning. However, the mechanism of learning is not known to us. Researchers around the world are trying to explore this mystery through developing various models and implementing them on various humanoid robots. All the models based on conventional mechanics and controls have inherent limitations. We believe appropriate computational model based on learning will be able to effectively address this issue. Accordingly, we have collected extensively humanoid push recovery data using our innovative idea of exploiting the accelerometer sensor of smart phone. Through our experiments, we have studied the human push recovery by fusing data at feature level using physics toolbar accelerometer of android interface kit. The subjects for the experiments were selected both as right handed and left handed. Pushes were induced from the behind with close eyes to observe the motor action as well as with open eyes to observe learning-based reactive behaviors. A learning vector quantization-based classifier has been developed to identify the coordination between various push and hip and knee joints.

Gait pattern performance, for its crucial significance in humanoid robot stabilization, has attracted the attention of researchers worldwide. Although simplified models highlight major features, bipedal walking has bewildered the researchers. Therefore, for a precise understanding of the bipedal model, a state-of-the-art, simplified model has been proposed in this paper which comprises a 3D-multilinked dual spring-loaded inverted pendulum (3D-MDSLIP) while acknowledging the vertical fluctuations of the center of mass (CoM). In addition, the model considers upper body movement and its effects on the stabilization of the humanoid robot. The mathematical modeling of a humanoid walking over the obstacle and slope is demonstrated to precisely understand the problem. The tuning process of the parameters and postures in a humanoid robot is complex and time-consuming. For proper walking of a robot over uneven terrains and slopes, tuning of the PID controller is achieved using converged teaching-learning based optimization (CTLBO) technique for a central pattern generator (CPG) gait, as introduced in the paper. The optimal gait angles are applied to the experimental and simulated NAO to successfully navigate the provided terrain. Thus, the experimental and simulation results jointly show that the proposed CPG-CTLBO gait learning technique is feasible for finding an optimal gait pattern for the humanoid robot within a deviation of 5%. The energy efficiency of the proposed controller is compared with the default controller of NAO based on the average electronic current in sagittal and lateral movement. Further, it is examined for the energy consumption for several slopes, and the results obtained are acceptable, showing the controller is efficient. Additionally, it has been compared with an existing technique for walking a humanoid robot on uneven terrains. The graph obtained using the proposed technique demonstrates the superiority of the proposed technique.

In this work, we present a novel method for ankle/foot compliance for biped humanoid robots walking over uneven terrain. Based on distributed plantar proximity sensing, we developed the Preemptive Foot Compliance (PFC) control that generates a Preemptive Ground Reaction Wrench that modifies the foot orientation to maximize the largest contact area to reduce the impact force produced at landing. PFC can be easily included in any walking controller for flat ground and become capable of walking on uneven terrains. The PFC was tested on two full-size humanoid robots running two different walking controllers, originally designed only for flat-ground walking. Both robots increase their capability to walk over uneven terrain and the walking impacts in flat grown were reduced by approximately 80%.

Specific problems were extracted in an experiment using a practical system in an attempt to transport an object cooperatively with two humanoid robots. The result proved that both body swinging during movement and the shift in the center of gravity, by transporting an object, caused a shift in the position after movement.