The target of design is expanding from manufactured products and services to society itself. Also the scope of the designer's role is spreading to a wider range of players throughout society. This means we must emphasize a new facet vital to the world of design – that of accepting and understanding differing viewpoints from the diverse range of members comprising society, and create a new system based on this new foundation. These phenomena show a change in the application of design, from an industrial and consumer orientation to a societal and democratic orientation. It also indicates a change in those responsible for design, from professional designers to people who generate and establish societal habits. This paper uses several arguments as a foundation to argue that our current societal processes generate various activities that can and should be used as environments for developing designs, rather than limiting work to the designer's studio. I call this concept “Societal Design.” Discussions in this paper are based on various aspects and phenomena realized during a workshop on welfare design in which the experimental co-design process was carried out by social activists and designers. The workshop was organized as an exploration of our research program on developing the Societal Design method and methodology.

Recently, the framework of shared control between a human operator and a robot system using a cooperative controller with human operator to decide control input to a robot system has been drawing much attention for dexterous telemanipulation. However, desiging a suitable controller for shared control scheme for the given task is difficult due to the difficulties in modeling of operator'’s behavior and environment. In this paper，we proposed a model-free approach using reinforcement learning to learn a shared control policy through interactions with the operator, robot and environment. To validate our method，we adopted a page turning task by telemanipulation and developed an experimental platform with a physical simulator．Experimental results suggest that our method is able to learn task-relevant shared control for flexible and enhanced dexterous manipulation．