A central problem in designing and implementing interactive systems---action selection---is also a core research topic in automated planning. While numerous toolkits are available for building end-to-end interactive systems, the tight coupling of representation, reasoning, and technical frameworks found in these toolkits often makes it difficult to compare or change the underlying domain models. In contrast, the automated planning community provides general-purpose representation languages and multiple planning engines that support these languages. We describe our recent work on automated planning for task-based social interaction, using a robot that must interact with multiple humans in a bartending domain.

Food and beverage manufacturers are still on the fence about artificial intelligence adoption. Many believe it isn't useful or plain do not know how to use it. According to research firm Mordor Intelligence, artificial intelligence in the food and beverage market was valued at U.S. $3.07 billion in 2020 and is expected to reach $29.94 billion by 2026 at a CAGR of over 45.77% during the forecast period, 2021-2026. The Association for Advancing Automation (A3) is made up of three principle daughter associations: robotics, machine vision and motion control/motor industries, where AI acts as a bridge across those technologies. One of the roles for Robert Huschka, vice president of Education Strategies, A3, is liaison to the strategic advisory committee on artificial intelligence.

As more manufacturers begin to integrate smart machines into their production processes, employees throughout the industry seek to understand what it will mean to the factory of the future. To understand how new industrial technology will impact employees, a team from Intel's Internet of Things Group conducted a study to find out what workers from the factory floor to the boardroom can expect in the intelligent factory of the future. Led by Dr. Irene Petrick, industrial innovation director, and Dr. Faith McCreary, principal engineer and researcher, the global study involved interviewing 145 employees working in industrial manufacturing -- including petrochemical, metal fabrication, and food and beverage companies with factories, primarily in North America. "Current manufacturers who are considering new digital implementations need to think not just about the technology piece, but also about their workforce," said Petrick, an internationally recognized expert in strategic road mapping ...

For robots to solve real world tasks, they often require the ability to reason about both symbolic and geometric knowledge. We present a framework, called KABouM, for integrating knowledge-level task planning and motion planning in a bounding geometry. By representing symbolic information at the knowledge level, we can model incomplete information, sensing actions and information gain; by representing all geometric entities-- objects, robots and swept volumes of motions--by sets of convex polyhedra, we can efficiently plan manipulation actions and raise reasoning about geometric predicates, such as collisions, to the symbolic level. At the geometric level, we take advantage of our bounded convex decomposition and swept volume computation with quadratic convergence, and fast collision detection of convex bodies. We evaluate our approach on a wide set of problems using real robots, including tasks with multiple manipulators, sensing and branched plans, and mobile manipulation.