Загрузка страницы

Dileep George at AI Frontiers 2017 : Opportunities and challenges for robot manipulation

Progress in artificial intelligence is starting to give robots the ability to perceive their environment. While deep learning has created classification systems that exceed human abilities in certain domains, manipulation tasks require a richer understanding of the world than classification or detection. To this end, Vicarious focuses on building systems with greater data efficiency, flexibility of reasoning, and transfer of knowledge between tasks. Combining task relevant features from deep learning with active perception, handling of uncertainty, and closed loop planning can mitigate many of the commonly encountered errors in robot manipulation. I'll describe our progress, the opportunities we are exploring, and the problems that remain to be solved.

Видео Dileep George at AI Frontiers 2017 : Opportunities and challenges for robot manipulation канала AI Frontiers
Показать
Комментарии отсутствуют
Введите заголовок:

Введите адрес ссылки:

Введите адрес видео с YouTube:

Зарегистрируйтесь или войдите с
Информация о видео
14 ноября 2017 г. 14:36:32
00:19:28
Яндекс.Метрика