Optimization of Central Pattern Generator-Based Torque-Stiffness-Controlled Dynamic Bipedal WalkingRead the full article
Journal of Robotics publishes original research articles as well as review articles on all aspects of automated mechanical devices, from their design and fabrication, to testing and practical implementation.
Journal of Robotics maintains an Editorial Board of practicing researchers from around the world, to ensure manuscripts are handled by editors who are experts in the field of study.
Latest ArticlesMore articles
Improved Manipulator Obstacle Avoidance Path Planning Based on Potential Field Method
Aiming at the existing artificial potential field method, it still has the defects of easy to fall into local extremum, low success rate and unsatisfactory path when solving the problem of obstacle avoidance path planning of manipulator. An improved method for avoiding obstacle path of manipulator is proposed. First, the manipulator is subjected to invisible obstacle processing to reduce the possibility of its own collision. Second, establish dynamic virtual target points to enhance the predictive ability of the manipulator to the road ahead. Then, the artificial potential field method is used to guide the manipulator movement. When the manipulator is in a local extreme or oscillating, the extreme point jump-out function is used in real time to make the end point of the manipulator produce small displacements and change the action direction to effectively jump out of the dilemma. Finally, the manipulator is controlled to avoid all obstacles and move smoothly to form a spatial optimization path from the start point to the end point. The simulation experiment shows that the proposed method is more suitable for complex working environment and effectively improves the success rate of manipulator path planning, which provides a reference for further developing the application of manipulator in complex environment.
Analysis on the Impact Factors for the Pulling Force of the McKibben Pneumatic Artificial Muscle by a FEM Model
Modelling the behaviour of Pneumatic Artificial Muscle (PAM) has proven difficult due to its highly complicated structure, nonlinear nature of rubbery material, and air compressibility. To overcome these limitations, a FEM (Finite Element Method) model using Abaqus and CATIA is derived for the quantitative analysis on the impact of different factors on the pulling force of PAM. In the Abaqus a two parameter Mooney–Rivlin model is utilized to consider the hyper-elastic nature of flexible material. Then both Abaqus and CATIA are used in the parametric design of a 3-Dimensional model of PAM. Furthermore, the FEM model is employed to predict the static force exerted by PAM and the results show that the model is promising. The FEM model produces closer results to the test data for the typical PAM. Nonlinear behaviour of PAM is found to be obvious with an increase in both the contraction and the air pressure, different from the linear curves obtained by the fundamental geometrical model. Nonlinear changes in the PAM force are also observed in the numerical study on the effect of structural factors including initial braid angle, initial diameter, initial wall thickness, and flexible material. Besides, these phenomena can be explained by a connection between mechanical and morphological behaviour of PAMs with the FEM model. Generally, this modelling approach is more accurate compared to the fundamental theoretical model and more cost competitive compared to the empirical methods.
Learning from Demonstrations and Human Evaluative Feedbacks: Handling Sparsity and Imperfection Using Inverse Reinforcement Learning Approach
Programming by demonstrations is one of the most efficient methods for knowledge transfer to develop advanced learning systems, provided that teachers deliver abundant and correct demonstrations, and learners correctly perceive them. Nevertheless, demonstrations are sparse and inaccurate in almost all real-world problems. Complementary information is needed to compensate these shortcomings of demonstrations. In this paper, we target programming by a combination of nonoptimal and sparse demonstrations and a limited number of binary evaluative feedbacks, where the learner uses its own evaluated experiences as new demonstrations in an extended inverse reinforcement learning method. This provides the learner with a broader generalization and less regret as well as robustness in face of sparsity and nonoptimality in demonstrations and feedbacks. Our method alleviates the unrealistic burden on teachers to provide optimal and abundant demonstrations. Employing an evaluative feedback, which is easy for teachers to deliver, provides the opportunity to correct the learner’s behavior in an interactive social setting without requiring teachers to know and use their own accurate reward function. Here, we enhance the inverse reinforcement learning () to estimate the reward function using a mixture of nonoptimal and sparse demonstrations and evaluative feedbacks. Our method, called from demonstration and human’s critique (), has two phases. The teacher first provides some demonstrations for the learner to initialize its policy. Next, the learner interacts with the environment and the teacher provides binary evaluative feedbacks. Taking into account possible inconsistencies and mistakes in issuing and receiving feedbacks, the learner revises the estimated reward function by solving a single optimization problem. The is devised to handle errors and sparsities in demonstrations and feedbacks and can generalize different combinations of these two sources expertise. We apply our method to three domains: a simulated navigation task, a simulated car driving problem with human interactions, and a navigation experiment of a mobile robot. The results indicate that the significantly enhances the learning process where the standard methods fail and learning from feedbacks () methods has a high regret. Also, the works well at different levels of sparsity and optimality of the teacher’s demonstrations and feedbacks, where other state-of-the-art methods fail.
Longitudinal Modeling and Control of Tailed Flapping-Wings Micro Air Vehicles near Hovering
Compared with the tailless flapping wing micro air vehicle (FMAV), the tailed FMAV has a simpler structure and is easier to control. However, although biplane FMAVs with tails have been used for flight control in practice for a long time, a theoretical model of the tailed FMAV has not previously been established. In this paper, we report modeling of the longitudinal dynamics of a tailed biplane FMAV using the Newton‐Euler equations. In this study, the vehicle was trimmed and linearized near its hovering equilibrium, assuming small disturbances. Then the stability of the hovering FMAV was analyzed with a modal analysis method. A state feedback controller was synthesized to stabilize the disturbance. Finally, we investigated the flight control of the tailed biplane FMAV with different control signals. Our results show that the natural‐motion mode determines the oscillation divergence characteristics of the tailed FMAV, a mode that can be suppressed with the state feedback controller by real‐time modulation of the tail. The tail can also be used to achieve different flight modes with different control‐signal functions. The tailed FMAV cruises in a line when the tail is controlled with a step function and spirals in an elliptical trajectory in the longitudinal plane when the tail is controlled by a sinusoidal function. Our longitudinal‐ dynamics model provides an analytical basis for further dynamic analyses of the tailed FMAV, as well as the corresponding controller synthesis. Moreover, the proposed attitude stabilization and flight control schemes for the vehicle near hovering provide a basis for developing practical uses of the tailed FMAV.
sEMG Based Human Motion Intention Recognition
Human motion intention recognition is a key to achieve perfect human-machine coordination and wearing comfort of wearable robots. Surface electromyography (sEMG), as a bioelectrical signal, generates prior to the corresponding motion and reflects the human motion intention directly. Thus, a better human-machine interaction can be achieved by using sEMG based motion intention recognition. In this paper, we review and discuss the state of the art of the sEMG based motion intention recognition that is mainly used in detail. According to the method adopted, motion intention recognition is divided into two groups: sEMG-driven musculoskeletal (MS) model based motion intention recognition and machine learning (ML) model based motion intention recognition. The specific models and recognition effects of each study are analyzed and systematically compared. Finally, a discussion of the existing problems in the current studies, major advances, and future challenges is presented.
Kinematic Analysis and Dynamic Optimization Simulation of a Novel Unpowered Exoskeleton with Parallel Topology
This paper studies the kinematic and dynamic analysis of a novel unpowered exoskeleton with topology. Firstly, the kinematics of the unpowered exoskeleton is analyzed by the derivation of the closed-loop position equation, and the forward position problems of the exoskeleton are obtained. Secondly, with the aim of doing some research in the dynamics, two of links for the exoskeleton are changed into flexible links. Some shapes concerning some parameters are acquired by simulation with fitting curve method. Thirdly, meanwhile, the dynamic model is built by using Lagrange method. Fourthly, the gait experiment is acquired with the aim of obtaining the law of the human joints. Fifthly, the dynamic model is verified by Adams software and the theoretical calculation. Meanwhile, an optimization is completed in the Adams software. The most reasonable spring stiffness is acquired. Finally, some conclusions are enumerated to show the properties of the mechanisms.