Autonomy of Mobile Robots: From Single to Multiple
He Yuqing, State Key Laboratory of Robotics, Shenyang Institute of Automation

In this presentation, some basic problems and demonstrations on the autonomy mobile robot systems are shown. They can be categorized as the following four aspects: 1) behavior environmental modeling, i.e., how a robot sense and process all the factors that may prevent it working as usual. In general, behavior environmental includes dynamics of robot itself, natural surroundings such as trees, stones, and so on, and the state of objects that the mobile robot may manipulate or cooperate with. This problem is usually very difficult because most of these factors are very complicated, highly nonlinear and dynamical. 2) Behavior decision, that is, how the robot itself can select some optimal or sub-optimal behavior, for example, continuous or discrete motion trajectory, structure transformation, with respect to the current behavior surroundings; 3) Ability of learning. In the aspect of learning, an intelligent or autonomous robot should have the following two abilities: one is that it should be capable of learning from human being, friend robots and even its own experiences to accumulate the so-called ‘knowledge’; the other is to use the so-called ‘knowledge’ to make a rapid decision-making. 4) Cooperation and coordination of multiple robots to improve the performance of a single robot. Finally, some existing general difficulties and basic problems in the field will be given to be made some further discussions with all the interested audience.

Reference:

[1] George Musser, The spirit of exploration, Scientific American, March 2004.

[2] Brenna D. Argall, Sonia Chernova, Manuela Veloso, Brett Browning, A survey of robot learning from demonstration, Robotics and Autonomous Systems, 57(5), 2009: 469-483.

Loading more stuff…

Hmm…it looks like things are taking a while to load. Try again?

Loading videos…