Search results

Human beings are able to adjust their arm stiffness in daily life tasks. This paper aims to enable a robot to learn these human-like variable stiffness motor skills autonomously.

The paper presents a reinforcement learning method to enable a robot to learn variable stiffness motor skills autonomously. Firstly, the variable stiffness motor skills are encoded by the previously proposed dynamical movement primitives and stiffness primitives (DMP-SP) framework, which is able to generate both motion and stiffness curves for robots. The admittance controller is then used to make a robot follow the motion and stiffness curves. The authors then use the policy improvement with path integrals (PI2) algorithm to optimize the robot motion and stiffness curves iteratively.

The performance of the proposed method is evaluated on an UR10 robot by two different tasks: a) via-point task, b) sweeping the floor. The results show that after training, the robot is capable of accomplishing the tasks safely and compliantly.

The method can help the robots walk out of the isolated environment and accelerate their integration into human being’s daily life.

This paper uses reinforcement learning method to improve DMP-SP framework, thus allowing the robots to learn variable stiffness motor skills autonomously with no need for extra sensors.

The purpose of this paper is to enable robots to intelligently adapt their damping characteristics and motions in a reactive fashion toward human inputs and task requirements during physical human–robot interaction.

This paper exploits a combination of the dynamical system and the admittance model to create robot behaviors. The reference trajectories are generated by dynamical systems while the admittance control enables robots to compliantly follow the reference trajectories. To determine how control is divided between the two models, a collaborative arbitration algorithm is presented to change their contributions to the robot motion based on the contact forces. In addition, the authors investigate to model the robot’s impedance characteristics as a function of the task requirements and build a novel artificial damping field (ADF) to represent the virtual damping at arbitrary robot states.

The authors evaluate their methods through experiments on an UR10 robot. The result shows promising performances for the robot to achieve complex tasks in collaboration with human partners.

The proposed method extends the dynamical system approach with an admittance control law to allow a robot motion being adjusted in real time. Besides, the authors propose a novel ADF method to model the robot’s impedance characteristics as a function of the task requirements.

The purpose of this paper is to present a method which enables a robot to learn both motion skills and stiffness profiles from humans through kinesthetic human-robot cooperation.

Admittance control is applied to allow robot-compliant behaviors when following the reference trajectories. By extending the dynamical movement primitives (DMP) model, a new concept of DMP and stiffness primitives is introduced to encode a kinesthetic demonstration as a combination of trajectories and stiffness profiles, which are subsequently transferred to the robot. Electromyographic signals are extracted from a human’s upper limbs to obtain target stiffness profiles. By monitoring vibrations of the end-effector velocities, a stability observer is developed. The virtual damping coefficient of admittance controller is adjusted accordingly to eliminate the vibrations.

The performance of the proposed methods is evaluated experimentally. The result shows that the robot can perform tasks in a variable stiffness mode as like the human dose in the teaching phase.

DMP has been widely used as a teaching by demonstration method to represent movements of humans and robots. The proposed method extends the DMP framework to allow a robot to learn not only motion skills but also stiffness profiles. Additionally, the authors proposed a stability observer to eliminate vibrations when the robot is disturbed by environment.

The purpose of this paper is to eliminate instability which may occur when a human stiffens his arms in physical human–robot interaction by estimating the human hand stiffness and presenting a modified vibration index.

Human hand stiffness is first estimated in real time as a prior indicator of instability by capturing the arm configuration and modeling the level of muscle co-contraction in the human’s arms. A time-domain vibration index based on the interaction force is then modified to reduce the delay in instability detection. The instability is confirmed when the vibration index exceeds a given threshold. The virtual damping coefficient in admittance controller is adjusted accordingly to ensure stability in physical human–robot interaction.

By estimating the human hand stiffness and modifying the vibration index, the instability which may occur in stiff environment in physical human–robot interaction is detected and eliminated, and the time delay is reduced. The experimental results demonstrate significant improvement in stabilizing the system when the human operator stiffens his arms.

The originality is in estimating the human hand stiffness online as a prior indicator of instability by capturing the arm configuration and modeling the level of muscle co-contraction in the human’s arms. A modification of the vibration index is also an originality to reduce the time delay of instability detection.

This paper aims to study the impact of manufacturer’s upgrading strategy of durable products on the retailer’s decision on trade-in program and her decision on the secondary market.

This paper develops a channel that consists of a manufacturer and a retailer, where the manufacturer releases an upgraded product, and the retailer introduces a trade-in program for consumers, simultaneously, decides whether to enter the secondary market. These approaches are modeled through Stackelberg game.

This paper reveals that the optimal conditions for manufacturer to release upgraded products and retailer to resell used products in the secondary market, and it reveals that under what conditions it is profitable for retailer to enter the secondary market under product upgrade levels.

If the manufacturer’s upgrade level is low, it is profitable for the retailer to enter the secondary market. However, if the manufacturer’s upgrade level is high, it is unprofitable for the retailer to enter the secondary market.

In this paper, the active secondary market, upgrading of new products, consumer market segmentation and especially, the upgrade degree of new products as a function of consumer demand are considered simultaneously.