What Can I Control?: The Development of Visual Categories for a Robot's Body and the World that it Influences (original) (raw)
Related papers
Sensorimotor coordination in a "baby" robot: learning about objects through grasping
Progress in brain research, 2007
This paper describes a developmental approach to the design of a humanoid robot. The robot, equipped with initial perceptual and motor competencies, explores the "shape" of its own body before devoting its attention to the external environment. The initial form of sensorimotor coordination consists of a set of explorative motor behaviors coupled to visual routines providing a bottom-up sensory-driven attention system. Subsequently, development leads the robot from the construction of a "body schema" to the exploration of the world of objects. The "body schema" allows controlling the arm and hand to reach and touch objects within the robot's workspace. Eventually, the interaction between the environment and the robot's body is exploited to acquire a visual model of the objects the robot encounters which can then be used to guide a top-down attention system.
What can I control? A framework for robot self-discovery
2006
In this paper we present a developmental progression for a humanoid robot that uses mutual information to discover controllable perceptual categories. Previously we have shown that the robot can discover a visual category that corresponds with its hand from less than 5 minutes of interaction with a human. Here, we show how this discovery can be used to adapt the robot's perceptual and motor systems such that the robot can subsequently discover its fingers. In this way, a robot can expand its control over the world so that newly mastered actions can open up new realms of influence, and new realms of influence can lead to the mastery of new actions.
Humanoid Learns to Detect Its Own Hands
Robust object manipulation is still a hard problem in robotics, even more so in high degree-of-freedom (DOF) humanoid robots. To improve performance a closer integration of visual and motor systems is needed. We herein present a novel method for a robot to learn robust detection of its own hands and fingers enabling sensorimotor coordination. It does so solely using its own camera images and does not require any external systems or markers. Our system based on Cartesian Genetic Programming (CGP) allows to evolve programs to perform this image segmentation task in real-time on the real hardware. We show results for a Nao and an iCub humanoid each detecting its own hands and fingers.
Sensorimotor processes for learning object representations
2007 7th IEEE-RAS International Conference on Humanoid Robots, 2007
Learning object representations by exploration is of great importance for cognitive robots that need to learn about their environment without external help. In this paper we present sensorimotor processes that enable the robot to observe grasped objects from all relevant viewpoints, which makes it possible to learn viewpoint independent object representations. Taking control of the object allows the robot to focus on relevant parts of the images, thus bypassing potential pitfalls of pure bottom-up attention and segmentation. We propose a systematic method to control a robot in order to achieve a maximum range of motion across the 3-D view sphere. This is done by exploiting the task redundancies typically found on a humanoid arm and by avoiding joint limits of the robot. The proposed method brings the robot into configurations that are appropriate for observing objects. It enables us to acquire a wider range of snapshots without regrasping the object.
From pixels to objects: Enabling a spatial model for humanoid social robots
2009 IEEE International Conference on Robotics and Automation, 2009
This work adds the concept of object to an existent low-level attention system of the humanoid robot iCub. The objects are defined as clusters of SIFT visual features. When the robot first encounters an unknown object, found to be within a certain (small) distance from its eyes, it stores a cluster of the features present within an interval about that distance, using depth perception. Whenever a previously stored object crosses the robot's field of view again, it is recognized, mapped into an egocentrical frame of reference, and gazed at. This mapping is persistent, in the sense that its identification and position are kept even if not visible by the robot. Features are stored and recognized in a bottom-up way. Experimental results on the humanoid robot iCub validate this approach. This work creates the foundation for a way of linking the bottom-up attention system with top-down, object-oriented information provided by humans.
Towards grasp-oriented visual perception for humanoid robots
2009
A distinct property of robot vision systems is that they are embodied. Visual information is extracted for the purpose of moving in and interacting with the environment. Thus, different types of perception-action cycles need to be implemented and evaluated.
Developing haptic and visual perceptual categories for reaching and grasping with a humanoid robot
Robotics and Autonomous Systems, 2001
Properties of the human embodiment — sensorimotor apparatus and neurological structure — participate directly in the growth and development of cognitive processes against enormous worst case complexity. It is our position that relationships between morphology and perception over time lead to increasingly comprehensive models that describe the agent’s relationship to the world. We are applying insight derived from neuroscience, neurology, and developmental psychology to the design of advanced robot architectures. To investigate developmental processes, we have begun to approximate the human sensorimotor configuration and to engage sensory and motor subsystems in developmental sequences. Many such sequences have been documented in studies of infant development, so we intend to bootstrap cognitive structures in robots by emulating some of these growth processes that bear an essential resemblance to the human morphology. In this paper, we will show two related examples in which a humanoid robot determines the models and representations that govern its behavior. The first is a model that captures the dynamics of a haptic exploration of an object with a dextrous robot hand that supports skillful grasping. The second example constructs constellations of visual features to predict relative hand/object postures that lead reliably to haptic utility. The result is a first step in a trajectory toward associative visual-haptic categories that bounds the incremental complexity of each stage of development.
Robots are gaining more attention in the neuroscientific community as means of verifying theoretical models of social skill development. In particular, humanoid robots which resemble dimensions of a child offer a controlled platform to simulate interactions between humans and infants. Such robots equipped with biologically inspired models of social and cognitive skill development might provide invaluable insights into learning mechanisms infants employ when interacting with others. One such mechanism which develops in infancy is the ability to share and direct attention of interacting participants. Pointing behaviour underlies joint attention and is preceded by hand-eye coordination. Here, we attempt to explain how pointing emerges from sensorimotor learning of hand-eye coordination in a humanoid robot. A robot learned joint configurations for different arm postures using random body babbling. Arm joint configurations obtained in babbling experiment were used to train biologically inspired models based on self-organizing maps. We train and analyse models with various map sizes and depending on their configuration relate them to different stages of sensorimotor skill development. Finally, we show that a model based on self-organizing maps implemented on a robotic platform accounts for pointing gestures when a human presents an object out of reach for the robot.
Exploiting Sensorimotor Coordination for Learning to Recognize Objects
2007
In this paper we present a system which learns to recognize objects through interaction by exploiting the principle of sensorimotor coordination. The system uses a learning architecture which is composed of reactive and deliberative layers. The reactive layer consists of a database of behaviors that are modulated to produce a desired behavior. In this work we have implemented and installed in our architecture an object manipulation behavior inspired by the concept that infants learn about their environment through manipulation. While manipulating objects, both proprioceptive data and exteroceptive data are recorded. Both of these types of data are combined and statistically analyzed in order to extract important parameters that distinctively describe the object being manipulated. This data is then clustered using the standard k-means algorithm and the resulting clusters are labeled. The labeling is used to train a radial basis function network for classifying the clusters. The performance of the system has been tested on a kinematically complex walking robot capable of manipulating objects with two legs used as arms, and it has been found that the trained neural network is able to classify objects even when only partial sensory data is available to the system. Our preliminary results demonstrate that this method can be effectively used in a robotic system which learns from experience about its environment.