Abstract
How do robots learn to perform motor tasks in a specific condition and apply what they have learned in a new condition? This paper proposes a framework for motor coordination acquisition of a robot drawing straight lines within a part of the workspace. Then, it addresses transferring the acquired coordination into another area of the workspace while performing the same task. Motor patterns are generated by a Central Pattern Generator (CPG) model. The motor coordination for a given task is acquired by using a multi-objective optimization method that adjusts the CPGs' parameters involved in the coordination. To transfer the acquired motor coordination to the whole workspace we employed (1) a Self-Organizing Map that represents the end-effector coordination in the Cartesian space, and (2) an estimation method based on Inverse Distance Weighting that estimates the motor program parameters for each SOM neuron. After learning, the robot generalizes the acquired motor program along the SOM network. It is able therefore to draw lines from any point in the 2D workspace and with different orientations. Aside from the obvious distinctiveness of the proposed framework from those based on inverse kinematics typically leading to a point-to-point drawing, our approach also permits of transferring the motor program throughout the workspace.
Highlights
The capacity of applying previously acquired skills in a new context is referred to as transfer of learning, e.g., the potential advantage of using the experience from a previously performed motor task to learn a new motor task
Where α0 = 1 is the slope of the sigmoid, ψ0 = 0 defines the center point. α represents the descending control from the high-level controller that modulates the activation of the pattern formation neuron pattern-formation layer (PF), which is shown in Figure 1 as αPF. wRG→PF is the weight of the synaptic connection between rhythm-generation layer (RG) and PF neurons
How can the coordination parameters be transferred from one starting position to another without learning from scratch? How can one obtain a representation for the motor coordination of the line drawing task over the workspace? We address the problem of transfer of motor coordination by employing an interpolation algorithm that uses previously acquired coordination parameters for an action in several arm configurations to estimate the coordination parameters for the same action in a new notpreviously-experienced arm configuration
Summary
The capacity of applying previously acquired skills in a new context is referred to as transfer of learning, e.g., the potential advantage of using the experience from a previously performed motor task to learn a new motor task. The robot movement is produced by learning non-linear differential equations in task space, a velocity-based inverse kinematics model is used to calculate the movement parameters in the joint space. Each line is drawn by a single CPG pattern generated at each joint, unlike inverse kinematics methods that perform the task by connecting a sequence of points. This has been achieved by a multi-objective optimization.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
