The importance of vision for robots is pervasive: from self-driving cars to detecting and handling objects for service robots in homes, from kitting in industrial workshops, to robots filling shelves and shopping baskets in supermarkets, etc. All these applications, and many more, imply interacting with a wide variety of objects, requiring in turn a deep understanding of what these objects look like, their properties, functionalities and likely locations.
There are robots performing complex tasks such as loading a dishwasher or flipping pancakes. However, the knowledge about the objects involved in these tasks is usually manually encoded within the robots control programs or knowledge bases, limiting them to operate on the objects they have been programmed to understand. This is not enough. Any robot, regardless of how much knowledge has been manually encoded into it, will inevitably face novel situations, and thus will always have gaps, conflicts or ambiguities in its own knowledge and capabilities. This calls for robots able to learn continuously about the objects they see by themselves.
The mission of Visual and Multimodal Applied Learning's (VANDAL) group (VANDAL) is to develop the body of theoretical knowledge and algorithms necessary to robots, and intelligent systems in general, to learn autonomously about objects in an open-ended manner. This implies using tools from machine learning, computer vision multimodal signal processing and analysis, data visualization and mining. Although intelligent embodied systems are the main application driving research in VANDAL, other applications are non-invasive control of prosthetic hands, scene understanding and automatic geolocalization.
