Organizers:
Abstract
As robots and autonomous systems move away from laboratory setups towards complex real-world scenarios, both the perception capabilities of these systems and their abilities to acquire and model semantic information must become more powerful. For example, a general-purpose service robot collaborating with a human user needs to know human spatial concepts and have an understanding of 3-dimensional objects, their use and functional relationships between them. More generally, semantic perception and mapping must become a resource for the robot, which links sensory information to the robot's knowledge base and high-level deliberative components. A key issue is the robot's ability to autonomously acquire and model the perceived semantic information, as well as (exploration) strategies for deciding where and how to acquire such information.
In this full-day workshop we will try to analyze the requirements of such a system, and discuss alternatives for achieving this goal, by bringing together researchers from areas such as Computer Vision, Cognitive Robotics, 3D Mapping, Mobile Manipulation, Machine Learning, Knowledge Representation, and Exploration. This workshop is planned to be the first event in a series of annual meetings at ICRA to discuss currently running projects and recent results to advance the state-of-the-art in semantic perception, mapping, and exploration.
Motivation and objectives:
Semantic perception for intelligent systems such as robots has seen a lot of progress recently, with many new and interesting techniques being developed in parallel by different research groups. To cope with a wide variety of tasks, service robots need to plan in spaces exceeding their local perceptual space. For deployment in human living environments, these robots need to aggregate and model perceived semantic information. A further requirement, for autonomous operation in the long-term, is the ability to acquire and update the necessary information, which involves actively exploring the environment.
A number of previous workshops have brought together researchers working on use of semantic information in a variety of domains including mapping for mobile robots, mobile manipulation, and semantic robot vision. The main focus of this workshop is on the autonomous acquisition of semantic information in intelligent robots and systems, as well as the use of semantic knowledge to guide further acquisition of information from the environment. The workshop will provide a common forum where people can exchange ideas and experiences on all aspects of this problem, and identify the common questions and concerns.
List of topics:
Topics of interest include, but are not necessary limited to:
Program Committee: (Those marked with an asterisk (*) have already agreed to contribute.)
Intended audience:
Researchers and R&D engineers working in the fields of personal
robotics, robot perception, mobile manipulation, 3D (semantic) mapping, and exploration. We also
hope to attract a large audience of student researchers (graduate
students) interested in the above mentioned topics. Based on related workshops,
we expect a peak attendance of 50-60 people.
Tentative schedule and agenda:
The workshop will be organized as a full-day workshop, with three paper sessions, and one poster session. The papers will be sorted into sessions based on their common research theme. After each paper session, there will be a brief panel discussion which aims to involve significant audience participation. Our goal is to link the workshop to previous events held at prestigious conferences in the mobile robotics field, including:
Relation to previous workshops: