• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.715-717
• /
• 2003

Some previous works for nearest neighbor (NN) query processing technique can treat a case that query/data are both moving objects. However, they cannot find exact result owing to vagueness of criterion. In order to escape their limitations and get exact result, we propose new NN query techniques, exact CTNN (continuous trajectory NN) query, approximate CTNN query, and dynamic CTNN query. These are all superior to pervious works, by reducing of number of calculation, considering of trajectory information, and using of continuous query concept. Using these techniques, we can solve any situations and types of NN query in location-aware environment.

• The Journal of the Korea Contents Association
• /
• v.17 no.8
• /
• pp.619-630
• /
• 2017

With the growing popularity of smart devices, various location based services have been providing to users. Recently, some location based social applications that combine social services and location based services have been emerged. The demands of a k-nearest neighbors(k-NN) query which finds k closest locations from a user location are increased in the location based social network services. In this paper, we propose an approximate k-NN query processing method for fast response time in a large number of users environments. The proposed method performs efficient stream processing using big data distributed processing technologies. In this paper, we also propose a modified grid index method for indexing a large amount of location data. The proposed query processing method first retrieves the related cells by considering a user movement. By doing so, it can make an approximate k results set. In order to show the superiority of the proposed method, we conduct various performance evaluations with the existing method.

• Journal of KIISE
• /
• v.42 no.2
• /
• pp.242-254
• /
• 2015

In this paper, we address the problem of improving the performance of multi-step k-NN search using multi-dimensional indexes. Due to information loss by lower-dimensional transformations, existing multi-step k-NN search solutions produce a large tolerance (i.e., a large search range), and thus, incur a large number of candidates, which are retrieved by a range query. Those many candidates lead to overwhelming I/O and CPU overheads in the postprocessing step. To overcome this problem, we propose two efficient solutions that improve the search performance by reducing the tolerance of a range query, and accordingly, reducing the number of candidates. First, we propose a tolerance reduction-based (approximate) solution that forcibly decreases the tolerance, which is determined by a k-NN query on the index, by the average ratio of high- and low-dimensional distances. Second, we propose a coefficient control-based (exact) solution that uses c k instead of k in a k-NN query to obtain a tigher tolerance and performs a range query using this tigher tolerance. Experimental results show that the proposed solutions significantly reduce the number of candidates, and accordingly, improve the search performance in comparison with the existing multi-step k-NN solution.

• International Journal of Control, Automation, and Systems
• /
• v.4 no.1
• /
• pp.17-29
• /
• 2006

A novel inverse kinematics solution based on the back propagation neural network (NN) for redundant manipulators is developed for online obstacles avoidance. A laser transducer at the end-effctor is used for online planning the trajectory. Since the inverse kinematics in the present problem has infinite number of joint angle vectors, a fuzzy reasoning system is designed to generate an approximate value for that vector. This vector is fed into the NN as a hint input vector rather than as a training vector to guide the output of the NN. Simulations are implemented on both three- and four-link redundant planar manipulators to show the effectiveness of the proposed position control system.

• KIISE Transactions on Computing Practices
• /
• v.21 no.11
• /
• pp.681-688
• /
• 2015

The k nearest neighbor (k-NN) graph construction is an important operation with many web-related applications, including collaborative filtering, similarity search, and many others in data mining and machine learning. Despite its many elegant properties, the brute force k-NN graph construction method has a computational complexity of $O(n^2)$, which is prohibitive for large scale data sets. Thus, (Key, Value)-based distributed framework, MapReduce, is gaining increasingly widespread use in Locality Sensitive Hashing which is efficient for high-dimension and sparse data. Based on the two-stage strategy, we engage the locality sensitive hashing technique to divide users into small subsets, and then calculate similarity between pairs in the small subsets using a brute force method on MapReduce. Specifically, generating a candidate group stage is important since brute-force calculation is performed in the following step. However, existing methods do not prevent large candidate groups. In this paper, we proposed an efficient algorithm for approximate k-NN graph construction by regrouping candidate groups. Experimental results show that our approach is more effective than existing methods in terms of graph accuracy and scan rate.

• Proceedings of the KIPE Conference
• /
• 2006.06a
• /
• pp.526-528
• /
• 2006

An adaptive output-feedback neural control problem of SISO strict-feedback nonlinear system is considered in this paper. The main contribution of the proposed method is that it is shown that the output-feedback control of the strict-feedback system can be viewed as that of the system in the normal form. As a result, proposed output-feedback control algorithm is much simpler than the previous backstepping-based controllers. Depending heavily on the universal approximation property of the neural network (NN) only one NN is employed to approximate lumped uncertain nonlinearity in the controlled system.

• Proceedings of the Korean Association of Geographic Inforamtion Studies Conference
• /
• 2010.06a
• /
• pp.53-60
• /
• 2010

위치 기반 서비스(Location-Based Services: LBS)에서 질의 요청자가 자신의 위치 정보와 원하는 질의를 전송하면, 위치 기반 서버는 이를 기반으로 질의를 처리하고 결과를 전송한다. 이 때 질의 요청자는 자신의 정확한 위치 좌표를 서버에 전송하기 때문에 개인 정보가 악용될 수 있는 위험에 노출된다. 이러한 문제를 해결하기 위하여 제안된 연구는 크게 Location Clocking 기법과 Private Information Retrieval(PIR) 기법으로 분류된다. Location Cloaking 기법은 사용자의 위치 좌표를 k-1개의 다른 사용자와 함께 묶어 하나의 Cloaking 영역을 생성하고 이를 바탕으로 질의를 처리한다. 그러나 영역에 대한 질의 후보 집합을 결과로 전송하므로 사용자에게 노출되는 POI 수가 증가하는 문제점을 지닌다. PIR은 암호화 기법으로 위치 기반 서버나 공격자에게 사용자의 위치와 질의 타입을 드러내지 않고 질의를 수행한다. 그러나 암호화 된 질의 결과로 사용자에게 데이터 전체를 전송하기 때문에 막대한 통신비용을 초래한다. 따라서 본 논문에서는 Location Cloakng과 PIR 기법의 장점을 결합하여 사용자의 개인 정보와 위치 기반 서버의 POI 정보 보호를 고려한 Approximate k-최근접점 질의 처리 알고리즘을 제안한다. 질의 전송시, 질의 요청자는 Cloaking 영역을 생성하여 위치 좌표를 감추고, 질의 결과 전송 시 Cloaking 영역에 제한된 PIR 프로토콜을 적용한다. 또한 k-최근접점 질의 수행시, 반환되는 POI의 수를 최소화하고, 정확도 높은 질의 결과를 만족하기 위해 Overlapping parameter를 적용한 색인 기법을 제안한다.

• 한국공간정보시스템학회:학술대회논문집
• /
• 2003.11a
• /
• pp.169-174
• /
• 2003

이동 객체에 대한 기존 최근접(nearest neighbor, NN) 질의 처리 기법들은 질의 궤적에 대해 연속적으로 정확하게, 질의와 가장 가까운 위치를 유지하면서 움직이는 최근접 객체를 선택할 수 있는 충분한 기준을 가지고 있지 못하다. 이 논문은 질의 객체와 데이터 객체가 모두 이동 객체인 경우에 가장 적합하게 사용되는 객체 궤적에 대한 연속적인 질의 처리를 통해 정확한 결과를 얻을 수 있는 새로운 최근접 질의 처리 기법, 연속 궤적 최근접 질의(CTNN, continuous trajectory nearest neighbor query)를 제안한다. 우리는 두 가지 Approximate, Exact CTNN 기법을 제안하며 이들은 모두 항해 시스템, 교통 통제 시스템, 물류정보 시스템 등 각종 위치 기반 서비스(L8S: location based services) 상에서 다양하게 사용될 수 있다. 이들은 이동 객체 궤적이 미리 알려져 있는 경우 그리고 질의와 데이터 객체가 모두 이동 객체인 경우에 가장 적합하다.

• Proceedings of the KIEE Conference
• /
• 1996.11a
• /
• pp.58-61
• /
• 1996

For a class of single-input single-output continuous-time nonlinear systems, a multilayer neural network-based controller that feedback-linearizes the system is presented. Control action is used to achieve tracking performance for a state-feedback linearizable but unknown nonlinear system. The multilayer neural network(NN) is used to approximate nonlinear continuous function to any desired degree of accuracy. The weight-update rule of multilayer neural network is derived to satisfy Lyapunov stability. It is shown that all the signals in the closed-loop system are uniformly bounded. Initialization of the network weights is straightforward.

• 제어로봇시스템학회:학술대회논문집
• /
• 2004.08a
• /
• pp.620-626
• /
• 2004

For general nonlinear processes, it is difficult to control with a linear model-based control method and nonlinear controls are considered. Among the numerous approaches suggested, the most rigorous approach is to use dynamic optimization. Many general engineering problems like control, scheduling, planning etc. are expressed by functional optimization problem and most of them can be changed into dynamic programming (DP) problems. However the DP problems are used in just few cases because as the size of the problem grows, the dynamic programming approach is suffered from the burden of calculation which is called as 'curse of dimensionality'. In order to avoid this problem, the Neuro-Dynamic Programming (NDP) approach is proposed by Bertsekas and Tsitsiklis (1996). To get the solution of seriously nonlinear process control, the interest in NDP approach is enlarged and NDP algorithm is applied to diverse areas such as retailing, finance, inventory management, communication networks, etc. and it has been extended to chemical engineering parts. In the NDP approach, we select the optimal control input policy to minimize the value of cost which is calculated by the sum of current stage cost and future stages cost starting from the next state. The cost value is related with a weight square sum of error and input movement. During the calculation of optimal input policy, if the approximate cost function by using simulation data is utilized with Bellman iteration, the burden of calculation can be relieved and the curse of dimensionality problem of DP can be overcome. It is very important issue how to construct the cost-to-go function which has a good approximate performance. The neural network is one of the eager learning methods and it works as a global approximator to cost-to-go function. In this algorithm, the training of neural network is important and difficult part, and it gives significant effect on the performance of control. To avoid the difficulty in neural network training, the lazy learning method like k-nearest neighbor method can be exploited. The training is unnecessary for this method but requires more computation time and greater data storage. The pH neutralization process has long been taken as a representative benchmark problem of nonlin ar chemical process control due to its nonlinearity and time-varying nature. In this study, the NDP algorithm was applied to pH neutralization process. At first, the pH neutralization process control to use NDP algorithm was performed through simulations with various approximators. The global and local approximators are used for NDP calculation. After that, the verification of NDP in real system was made by pH neutralization experiment. The control results by NDP algorithm was compared with those by the PI controller which is traditionally used, in both simulations and experiments. From the comparison of results, the control by NDP algorithm showed faster and better control performance than PI controller. In addition to that, the control by NDP algorithm showed the good results when it applied to the cases with disturbances and multiple set point changes.