No abstract

No abstract

There are many cases that a neural-network-based system must memorize some new patterns incrementally. However, if the network learns the new patterns only by referring to them, it probably forgets old memorized patterns, since parameters in the network usually correlate not only to the old memories but also to the new patterns. A certain way to avoid the loss of memories is to learn the new patterns with all memorized patterns. It needs, however, a large computational power. To solve this problem, we propose incremental learning methods with retrieving interfered patterns (ILRI). In these methods, the system employs a modified version of a resource allocating network (RAN) which is one variation of a generalized radial basis function (GRBF). In ILRI, the RAN learns new patterns with a relearning of a few number of retrieved past patterns that are interfered with the incremental learning. In this paper, we construct ILRI in two steps. In the first step, we construct a system which searches the interfered patterns from past input patterns stored in a database. In the second step, we improve the first system in such a way that the system does not need the database. In this case, the system regenerates the input patterns approximately in a random manner. The simulation results show that these two systems have almost the same ability, and the generalization ability is higher than other similar systems using neural networks and k-nearest neighbors.

Vision-based myoelectric prosthetic hand uses a camera integrated into its body for object detection and environment understanding, where the results provide necessary information for grasp planning. It is expected that the semi-automatic prosthesis control can be realized with this method. However, such a control method usually suffers from heavy computation due to the requirement of real-time image processing to keep up with the arm movements of the user. This paper presents a distributed control system that assigns heavy processing tasks to one or multiple processing nodes through the network, which greatly reduces the computation burdens of the processor embedded in the prosthetic hand. In this control scheme, the embedded system in the prosthetic hand is only used for gathering necessary data for grasp planning, while the processing nodes in the network are responsible for processing and managing the collected data. A test platform is built to verify the proposed control scheme. The test platform streams user electromyography (EMG) signals and images simultaneously to the GPU server. The GPU sever analyzes the received data and generates the corresponding motor commands in real time. A case study that uses a 3-DoF gripper to continuously grasp several objects is performed using this test platform.INDEX TERMS Distributed control system, real-time object detection, vision-based myoelectric hand.

The k-nearest neighbor (KNN) classification is a simple and effective classification approach. However, improving performance of the classifier is still attractive. Combining multiple classifiers is an effective technique for improving accuracy. There are many general combining algorithms, such as Bagging, Boosting, or Error Correcting Output Coding that significantly improve the classifier such as decision trees, rule learners, or neural networks. Unfortunately, these combining methods developed do not improve the nearest neighbor classifiers. In this paper, jirst, we present a new approach to combine multiple KNN classi-Jiers based on different distance functions, in which we apply multiple distance functions to improve the performance of the k-nearest neighbor classifiel: Second, we develop a combining method, in which the weights of the distance function, are learnt by genetic algorithm. Finally, combining classifiers in error correcting output coding, are discussed. The proposed algorithms seek to increase generalization accuracy when compared to the basic k-nearest neighbor algorithm. Experiments have been conducted on some benchmark datasets from the UCI Machine Learning Repository. The results show that the proposed algorithms improve the performance of the k-nearest neighbor classification.