Abstract:Purpose Under the development trend of intelligent manufacturing, the unstructured environment requires the robot to have a good generalization performance to adapt to the scene changes. The purpose of this paper aims to present a learning from demonstration (LfD) method (task parameterized [TP]-dynamic movement primitives [DMP]-GMR) that combines DMPs and TP-LfD to improve generalization performance and solve object manipulation tasks. Design/methodology/approach The dynamic time warping algorithm is applied to processing demonstration data to obtain a more standard learning model in the proposed method. The DMPs are used to model the basic trajectory learning model. The Gaussian mixture model is introduced to learn the force term of DMPs and solve the problem of learning from multiple demonstration trajectories. The robot can learn more local geometric features and generalize the learned model to unknown situations by adding task parameters. Findings An evaluation criterion based on curve similarity calculated by the Frechet distance was constructed to evaluate the model’s interpolation and extrapolation performance. The model’s generalization performance was assessed on 2D virtual data sets, and first, the results show that the proposed method has better interpolation and extrapolation performance than other methods. Originality/value The proposed model was applied to the axle-hole assembly task on real robots, and the robot’s posture in grasping and placing the axle part was taken as the task parameter of the model. The experiment results show that The proposed model is competitive with other models.

Enhanced Task Parameterized Dynamic Movement Primitives by GMM to Solve Manipulation Tasks

COMBINATION OF AFFINE DEFORMATION AND DYNAMIC MOVEMENT PRIMITIVE IN LEARNING HUMAN MOTION FOR REDUNDANT MANIPULATOR

Learning from demonstration using improved dynamic movement primitives

The Art of Imitation: Learning Long-Horizon Manipulation Tasks from Few Demonstrations

Combination of Dynamical Movement Primitives with Trajectory Segmentation and Node Mapping for Robot Machining Motion Learning

A Constrained DMPs Framework for Robot Skills Learning and Generalization From Human Demonstrations

Logic Learning from Demonstrations for Multi-step Manipulation Tasks in Dynamic Environments

Research and Implementation of Complex Task Based on DMP

Phase-free Dynamic Movement Primitives Applied to Kinesthetic Guidance in Robotic Co-manipulation Tasks

A Generalized Learning Method for Obstacle Constraint Tasks Based on Dynamic Movement Primitives

DMP and GMR Based Teaching by Demonstration for a KUKA LBR Robot

Extended DMPs Framework for Position and Decoupled Quaternion Learning and Generalization

Imitating Via Manipulability: Geometry-aware Combined DMP with Via-Point and Speed Adaptation

Overcoming Some Drawbacks of Dynamic Movement Primitives

A Task-Learning Strategy for Robotic Assembly Tasks from Human Demonstrations

Improved Dynamic Movement Primitive: A Method to Improve Trajectory Endpoint Convergence

Generalization of Task Parameterized Dynamical Systems using Gaussian Process Transportation

An improved approach of task-parameterized learning from demonstrations for cobots in dynamic manufacturing

Dynamic Movement Primitives Based Robot Skills Learning

A DMP-based Framework for Efficiently Generating Complete Stiffness Profiles of Human-like Variable Impedance Skill from Demonstrations