Typical mobile robots can be customized to perform a variety of different tasks by combining in different ways a set of basic control modules, or behaviors. By contrast, most current systems for manipulation are still designed for just one specific task. In this paper, we propose a hierarchical behavior-based system that can perform several vision-based manipulation tasks by using different combinations of the same set of basic behaviors. Behaviors can run concurrently, and they are arbitrated through "if-then" rules. We show experiments involving object tracking, grasping and placing, both with static and moving objects.

Behavior-based approaches to robot control are extremely popular in mobile robotics, but still rarely used for manipulation. We propose a behavior-based system that performs manipulation tasks using visual feedback. The distinctive points of our proposal are: (i) visual behavior is implemented using a new camera-based approach; (ii) reactive fuzzy rules are used to arbitrate behavior; and (iii) the outputs of concurrent behavior are fused using fuzzy logic. We show experiments on a real arm performing a pick-and-place task that illustrate our approach and demonstrate its advantages over current approaches.

We describe a new technique for landmark-based self-localization which is suitable for robots with poor odometry. This technique uses fuzzy logic to account for errors and imprecision in visual recognition, and for extreme uncertainty in the estimate of the robot's motion. It only requires an approximate model of the sensor system and a qualitative estimate of the robot's displacement, and it has a moderate computational cost. We show examples of use of our technique on a Sony AIBO legged robot in the RoboCup domain.