by Andrey Kolobov
by Fahiem Bacchus
by Alessandro Cimatti
by Maria Fox and Daniele Magazzeni
Background and State of the Art. Hybrid systems are systems with both continuous control variables and discrete logical modes. Many interesting real problems are indeed hybrid, including logistics planning, mission planning for autonomous vehicles, oil refinery management, supply management and disaster recovery. In these domains there are temporal, spatial and continuous constraints, and planning techniques capable of reasoning with suitably rich models are required. This motivates both increases in the modelling power to capture these constraints and extensions to classical approaches to planning to enable efficient management of such complex problems. This tutorial provides an introduction to modelling and solving planning problems in hybrid domains.
Recent Advances and New Challenges. This tutorial presents new benchmark problems motivating the need for modelling more realistic domains, and presents recent advances in the interaction between planning and robot control systems, aiming to generate sensible plans for real world situations. The tutorial ends with a discussion on current challenges and open problems in planning with hybrid domains.
by Andrew Coles and Luke Hunsberger
by Gabriele Röger and Malte Helmert
by Manuela Veloso
The creation of intelligent autonomous robots has been the research focus on many AI researchers. An autonomous robot perceives its environment, selects actions to achieve its goals based on the perceived state of the environment, and then executes the selected actions. This overall cycle is clearly of great relevance to the planning research community.
Symbiotic-autonomous robots are intelligent robots that are capable of complementing their perceptual, reasoning, and actuation capabilities by resourcing to external sources, such as humans, the web, and other artifacts. The robots further learn from their symbiotic interactions with the goal of improving their future tasks.
In this tutorial, I will introduce the core planning challenges for such symbiotic-autonomous robots, and present effective solutions for the underlying task-based planning, execution, and learning. The tutorial will be based on the concrete experience with our CoBot service mobile robots, which move in our multi-floor building performing a variety of pick and delivery tasks. The CoBot robots plan for their navigation and their interaction with humans and with the web. They have moved in our buildings for more than 200km.
by Holger H. Hoos
10th–14th June 2013