Object recognition in actual indoor environment is difficult at present even if using Deep Learning method, in which illumination condition greatly changes, and many objects are randomly put; even they may be physically overlapped. In this research, an object recognition method for robot to grasp objects is proposed, which makes the best use of robot capability of actively moving. The method is concretely as follows: small light and web-camera are set on the tip of robot arm. The robot actively illuminates the object (Active Lighting), actively changes the viewpoint of web-camera (Active Vision), and pushes and/or slides the overlapped objects to separate them (Graspless Manipulation). The effectiveness of proposed method was experimentally verified in case of 50 complicated scenes using a home robot equipped with dual arms and vision sensors.
A recurrent neural network (RNN) is a class of deep neural networks and able to imitate the behavior of dynamical systems due to its feedback mechanism. However, the feedback mechanism may cause network instability and hence the stability analysis of RNNs has been an important issue. From control theoretic viewpoint, we can readily apply the small gain theorem for the stability analysis of an RNN by representing it as a feedback connection with a linear time-invariant (LTI) system and a static nonlinear activation function typically being a rectified linear unit (ReLU). It is nonetheless true that the standard small gain theorem leads to conservative results since it does not care the important property that the ReLU returns nonnegative signals only. This motivates us to analyze the L2 induced norm of LTI systems for nonnegative input signals, which is referred to the L2+ induced norm in this paper. We characterize an upper bound of the L2+ induced norm by copositive programming, and then derive a numerically tractable semidefinite programming problem for (loosened) upper bound computation. We finally derive an L2+-induced-norm-based small gain theorem for the stability analysis of RNNs and illustrate its effectiveness by numerical examples.