• Journal of Institute of Control, Robotics and Systems
• /
• v.17 no.12
• /
• pp.1256-1265
• /
• 2011

Automated parking assist systems are being commercialized and rapidly spread in the market. In order to improve odometry accuracy, we proposed a practical odometry calibration scheme of Car-Like Mobile Robot (CLMR). However, there were some open problems in our prior work. For example, it was not clear whether the kinematic parameters always converged or not using the proposed calibration scheme. In addition, test driving had to be carried out "twice" without detailed explanation. This research aims to provide answers for the addressed questions though the convergence property analysis of the calibration scheme. In this paper, we evaluate on the effect of the kinematic parameter error on the odometry error at the final pose by numerical computation. The evaluation will show that the wheel diameter and tread of the CLMR can be calibrated by iterative test drives. In addition, the region of convergence in the parametric space will be discussed. Presented experimental results clearly showed that the proposed calibration scheme would be useful in practical applications.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.16 no.4
• /
• pp.153-163
• /
• 2017

This research is based on the previous research that personally preferred safe distance, rotating angle and speed are differentiated. Thus, we use machine learning model for recognizing maneuvering modes trained per personal or per similar driving pattern groups, and we evaluate automatic driving according to maneuvering modes. By utilizing driving knowledge, we subdivided 8 kinds of longitudinal modes and 4 kinds of lateral modes, and by combining the longitudinal and lateral modes, we build 21 kinds of maneuvering modes. we train the labeled data set per time stamp through RNN, LSTM and Bi-LSTM models by the trips of drivers, which are supervised deep learning models, and evaluate the maneuvering modes of automatic driving for the test data set. The evaluation dataset is aggregated of living trips of 3,000 populations by VTTI in USA for 3 years and we use 1500 trips of 22 people and training, validation and test dataset ratio is 80%, 10% and 10%, respectively. For recognizing longitudinal 8 kinds of maneuvering modes, RNN achieves better accuracy compared to LSTM, Bi-LSTM. However, Bi-LSTM improves the accuracy in recognizing 21 kinds of longitudinal and lateral maneuvering modes in comparison with RNN and LSTM as 1.54% and 0.47%, respectively.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.17 no.5
• /
• pp.188-199
• /
• 2018

In this study a novel dynamic lane-level location referencing method(LLRM) was developed. The terminologies were defined and the prerequisites were suggested for the LLRM. Then, the logical and physical data formats were proposed, followed by the development of encoding and decoding algorithms. To conduct a performance test of the proposed method, two different high precision digital maps were prepared as well as an evaluation tool. The test results demonstrated that the proposed method works perfectly in terms of accuracy. The processing speed and the data size were found to be less efficient, but it is expected that the defect would be compensated soon enough due to the fast growing technology of ICT and computer hardwares.