Abstract
Improvisation is a hallmark of human creativity and serves a functional purpose in completing everyday tasks with novel resources. This is particularly exhibited in tool-using tasks: When the expected tool for a task is unavailable, humans often are able to replace the expected tool with an atypical one. As robots become more commonplace in human society, we will also expect them to become more skilled at using tools in order to accommodate unexpected variations of tool-using tasks. In order for robots to creatively adapt their use of tools to task variations in a manner similar to humans, they must identify tools that fulfill a set of task constraints that are essential to completing the task successfully yet are initially unknown to the robot. In this paper, we present a high-level process for tool improvisation (tool identification, evaluation, and adaptation), highlight the importance of tooltips in considering tool-task pairings, and describe a method of learning by correction in which the robot learns the constraints from feedback from a human teacher. We demonstrate the efficacy of the learning by correction method for both within-task and across-task transfer on a physical robot.
Highlights
The abundant use of tools for a large range of tasks is a hallmark of human cognition (Vaesen, 2012)
While a robot may learn to complete a new task with a new tool via demonstrations by a human teacher (Argall et al, 2009; Rozo et al, 2013), the demonstration(s) provided for that tool cannot prepare the robot for all variations of that tool it is likely to encounter
Low task performance may be caused by the tooltip no longer contacting any relevant objects in the task, or by collisions between the tool’s new configuration and its environment that prevent the robot from executing the full trajectory
Summary
The abundant use of tools for a large range of tasks is a hallmark of human cognition (Vaesen, 2012). Prior work in creative robotics has often fallen under one of two categories of creativity: 1) Producing a creative output involving creative domains such as music (Gopinath and Weinberg, 2016) and painting (Schubert and Mombaur, 2013), or 2) Invoking a creative reasoning process. Within the latter category, several criteria for creative reasoning have been proposed, such as autonomy and self-novelty (Bird and Stokes, 2006), in which the robot’s creative output is novel to itself but not necessarily to an outside observer. In the context of a robot reasoning over how it may execute a task in a new environment, this co-creative process allows the robot to obtain the contextual knowledge needed to adapt its task model to meet the constraints of the novel environment
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
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 callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
