An online semi-supervised active learning algorithm is proposed, which is based on self-organizing incremental neural network (SOINN). The proposed method do not need any priori knowledge such as number of nodes or nu...
详细信息
As the computer data storage capacity increases and technology of digital imaging progresses rapidly, today we can access and manipulate huge image database on the Internet. How to search and access intelligently on c...
详细信息
Introductory programming course have two very specific difficulties for novice students. First is the lack of real world examples in the sessions. It is very difficult to find areas of application where all the studen...
详细信息
ISBN:
(纸本)9781595938657
Introductory programming course have two very specific difficulties for novice students. First is the lack of real world examples in the sessions. It is very difficult to find areas of application where all the students are familiar enough and that offers challenging and engaging examples. Second it is the lack of palpable results of the job done. Introductory courses in other fields generate products that the students can show to others, and feel proud about it. In CS1, for example, explaining loops by printing a series of numbers on the screen doesn't yield the same sense of accomplishment as drawing a basic perspective in an architecture course. We propose using Project Hoshimi [1], a Microsoft Platform, as a base for introducing computer programming to CS1 students. Through the paper we discuss how it could be implemented in the classroom, the main advantages and disadvantages in our experience of using Project Hoshimi, comparing its use against other more traditional approaches, as well as against other graphic programming methods such as Alice or videogame based learning.
We present a general framework for a higher-order spline level-set (HLS) method and apply this to smooth surface constructions. Starting from a first order energy functional, we obtain a general level set formulation ...
详细信息
Frequent localization in sensor networks may be needed due to the dynamically changing topology and the possible mobility of sensor nodes. We present a distributed adaptive localization method that we refer to as: map...
详细信息
Frequent localization in sensor networks may be needed due to the dynamically changing topology and the possible mobility of sensor nodes. We present a distributed adaptive localization method that we refer to as: map-based adaptive positioning (MAP). The main idea is to construct a relative local map at every node in the network, consisting of the node itself and its immediate neighbors, and merge the local maps together to form a global map. We consider two algorithms that can be used to estimate the relative local maps: multidimensional scaling (MDS) and semidefinite programming (SDP). The performance of these algorithms depend on two parameters: size of a local map, i.e., number of nodes, and the average connectivity of the node at the center of the local map and its 1-hop neighbors. We use machine learning to adaptively select the appropriate algorithm to estimate the relative local maps. Simulation results show that MAP outperforms both MDS and SDP, with better improvement for networks with less uniform node deployment.
This paper concerns a swing-up control problem for a 3-link gymnastic planar robot in a vertical plane, whose first joint is passive (unactuated) and the rest are active (actuated). The objectives of this paper are: 1...
详细信息
ISBN:
(纸本)9781424414970;1424414970
This paper concerns a swing-up control problem for a 3-link gymnastic planar robot in a vertical plane, whose first joint is passive (unactuated) and the rest are active (actuated). The objectives of this paper are: 1) to design a controller under which the robot can be brought into any arbitrarily small neighborhood of the upright equilibrium point, where all three links of the robot remain in their upright positions;2) to attain a global analysis of the motion of the robot under the controller. To tailor the energy based control approach to achieve the above objectives, first, this paper considers the links 2 and 3 as a virtually composite link and proposes a coordinate transformation of the angles of active joints. Second, this paper constructs a novel Lyapunov function based on the transformation, and devises an energy based swing-up controller. This paper presents a necessary and sufficient condition for nonexistence of any singular points in the controller for the robot starting from any initial state. Third, this paper carries out a global analysis of the motion of the robot under the controller, and establishes some conditions on control parameters for achieving the swing-up control objective. To validate the theoretical results obtained, this paper provides simulation results for a 3-link robot whose mechanical parameters are obtained from a human gymnast.
Requirements traceability is becoming increasingly significant element in software engineering. It provides critical function in the development and maintenance of a software system. From the software evolution point ...
详细信息
Requirements traceability is becoming increasingly significant element in software engineering. It provides critical function in the development and maintenance of a software system. From the software evolution point of view, requirements traceability plays an important role in facilitating software evolution. Since the evolution is inevitable, a traceability approach must take as much as possible the important influencing aspects into account to the evolution processes in order to minimize the evolution efforts. This paper evaluates several recent traceability approaches published in literature with the focus on their contributions to software evolution. The evaluation results may be used as a basis for improving requirements traceability approaches that may simplify the software evolution tasks.
Noise reduction is an important image processing method which has wide applications in different fields. The key to noise reduction is to reduce the noise without deteriorating the important features in the images. An...
详细信息
This paper presents a new particle method, with stochastic parameter estimation, to solve the SLAM problem. The underlying algorithm is rooted on a solid probabilistic foundation and is guaranteed to converge asymptot...
详细信息
This paper presents a new particle method, with stochastic parameter estimation, to solve the SLAM problem. The underlying algorithm is rooted on a solid probabilistic foundation and is guaranteed to converge asymptotically, unlike many existing popular approaches. Moreover, it is efficient in storage and computation. The new algorithm carries out filtering only in the marginal filtering space, thereby allowing for the recursive computation of low variance estimates of the map. The paper provides mathematical arguments and empirical evidence to substantiate the fact that the new method represents an improvement over the existing particle filtering approaches for SLAM, which work on the joint path state space.
暂无评论