Iterative affordance learning with adaptive action generation

Abstract

Abstract-A robot designer can provide a robot with knowledge to perform tasks on an environment. However, this approach can limit the achievement of future tasks executed by the robot. Providing it with the ability to develop its own skills paves the way for robots that are not limited by design. In this work a task consists in reproducing a given set of effects on an object. A robot must accomplish this task with limited information about the object, learning affordances to reproduce the effects, increasing this information throughout consecutive interactions with the object. We propose a method named Adaptive Affordance Learning (A 2 L) which endows a robot with the capacity to learn affordances associated to an object, both adapting the robot's actions to the object position, and increasing the robot's information about the object when needed. This paper presents two main contributions: first, an online adaption of the robot actions to interact with the object, decomposing each action into a sequence of movements, adapting each movement, in a close loop, to the object position; and second, to increase the information about the object, we propose an iterative process that alternates between (1) exploration of the environment interacting with the object, (2) affordance acquisition and (3) affordance validation. These contributions are assessed in two experiments where a simulated Baxter robot learns to push a box to different positions on a table.

Action Generation Adapted to Low-Level and High-Level Robot-Object Interaction States

“…The skill was implemented as a Bayesian Network (BN). In our previous work (Maestre et al, 2017) we identified a task-agnostic BN structure useful for the action generation. Therefore, building the skill consists in learning the Conditional Probabilistic Distributions (CPDs) of the BN.…”

“…In the current work, the structure represents the knowledge that an interaction is based on the relative position of the end-effector and an object, and the actual values of the interaction are stored as CPDs. In our previous work (Maestre et al, 2017) a simulated Baxter robot executed an exploration of a static environment identifying the BN structure and CPDs to push an object in different directions. The results demonstrated that the BN structure is task- and environment-agnostic.…”

“…Based on this principle, in our previous work (Maestre et al, 2017) we developed a method for a robot to build its own skills. A simulated Baxter robot endowed with our method executed an exploration of a static environment learning to push an object to specific positions of the environment.…”

Beyond the Self: Using Grounded Affordances to Interpret and Describe Others’ Actions

“…Robots operating in the real, unstructured world may understand available opportunities conditioned on their body, perception and sensorimotor experiences: the intersection of these elements gives rise to object affordances (action possibilities), as they are called in psychology [37]. The advantage of robot affordances lies in the ability to capture essential functional properties of environment objects in terms of the actions that the agent is able to perform with them, allowing to reason with prior knowledge about never-before-seen scenarios, thus exhibiting learning [38], [39] and some degree of online adaptation [40].…”