This study describes driving word2vec (DW2V), a new method for forming semantic representations of naturalistic driving data (NDD). To use big NDD for developing driver assistance systems or other information services, it is important to compress large amounts of data into an abstract and compact representation without losing semantic information. For this purpose, this study uses a symbolization method using a double articulation analyzer (DAA) assuming that NDD and human speech signals share an analogous structure, called a double articulation structure. The DAA can encode driving behavior data into sequences of driving words. However, the amount of semantic information contained in these sequences has not been clarified. Very few attempts have been made to develop a method for obtaining an adequate semantic representation of driving words that explains the relationship between different driving words. DW2V uses word2vec, proposed by Mikolov et al., to make a system learn the distributed semantic vector representation of symbolized naturalistic driving data (SNDD). Through experiments, we show that DW2V can restore the semantic relationships between different driving scenes from only a set of sequences of driving words, i.e., SNDD. In addition to quantitative analysis, a qualitative analysis of DW2V and its potential applications are discussed.

In this paper, we propose a self-localization method that exploits object recognition results by using convolutional neural networks (CNNs) for autonomous vehicles. Monte-Carlo localization (MCL) is one of the most popular localization methods that use odometry and distance sensor data for determining vehicle position. Some errors are often observed in the localization tasks and MCL often suffers from global positional errors. A global positional error means that particles representing a vehicle's position are distributed in the form of a multimodal distribution, i.e., the distribution has several peaks. To overcome this problem, we propose a method in which an autonomous vehicle employs object recognition results, obtained using CNNs, as the measurement data with a Bag-of-Features representation in an integrative manner. The semantic information found in the recognition results obtained using the CNN reduces the global errors in localization. The experimental results show that the proposed method can converge the distribution of the vehicle positions and particle orientations and reduce the global positional errors.

Driving behavioral data is too high-dimensional for people to review their driving behavior. It includes accelerator opening rate, steering angle, brake Master-Cylinder pressure and other various information. The high-dimensional data is not very intuitive for drivers to understand their driving behavior when they take a look back on their recorded driving behavior. We used a deep sparse autoencoder to extract the low-dimensional high-level representation from high-dimensional raw driving behavioral data obtained from a control area network. Based on this low-dimensional representation, we propose two visualization methods called Driving Cube and Driving Color Map. Driving Cube is a cubic representation displaying extracted three-dimensional features. Driving Color Map is a colored trajectory shown on a road map representing the extracted features. The trajectory is colored using the RGB color space, which corresponds to the extracted three-dimensional features. To evaluate the proposed method for extracting low-dimensional feature, we conducted an experiment and found several differences with recorded driving behavior by viewing the visualized Driving Color Map and that our visualization methods can help people to recognize different driving behavior. To evaluate the effectiveness of low-dimensional representation, we compared deep sparse autoencoder with other conventional methods from the viewpoint of linear separability of elemental driving behavior. As a result, our methods outperformed other conventional methods.

Computer Games are extremely challenging benchmarks for artificial intelligence in general. They need dynamic path planning in dynamic environments, learning ability and cooperative behaviors. We focus on Mario AI benchmark that evaluates AI controllers which is made by participants. To challenge the benchmark, we discuss the cooperative intelligence. It is difficult that agents obtain an optimal action rule in Mario AI benchmark because of large state space and time limit and so on. In this article, we add learning ability to search algorithm. It would be one kind of cooperative intelligence. We focus on A* algorithm for Search algorithm and Q-learning for learning ability. By some experiments, we show that the proposed method works well in MarioAI.

We propose two new generation methods of personal dictionary for handwritten character recognition using the set of characters written by a similar writer. The methods employ only one character written by one specific writer, and its character selects the set of characters written by the similar writer to generate the personal dictionary for the specific writer. The first type (similar mean) dictionary uses the mean feature vector of a similar writer which is selected by only one specific writer's character. The second type (similar feature space) dictionary uses the mean feature vector and the covariance matrix of the selected similar writer. We compared the effect for handwritten Japanese “HIRAGANA” characters. The similar feature space dictionary obtained the recognition rate 91% relatively to the rate of a general dictionary 82 %. It is confirmed that only one character by a specific writer is very effective on personal character recognition.

A new 17-GHz radio interferometer dedicated for solar observations was constructed in two years at Nobeyama, Nagano. It consists of eighty-four 80-cm-diameter antennas arranged in a tee-shaped array extending 490 m in east-west and 220 m in north-south directions. Since late June of 1992, radio full-disk images of the Sun have been observed for 8 h every day. The spatial resolution is 10" and the temporal resolution is 1 s and also 50 ms for selected events. Every 10 s correlator data are synthesized into images in real time and displayed on a monitor screen. The array configuration is optimized to observe the whole Sun with high spatial and temporal resolution and a high dynamic range of images. Image quality of better than 20 dB is realized by incorporation of technical advances in hardware and software, such as (1) low-loss phase-stable optical-fiber cables for local reference signal and IF signals, (2) newly developed phase-stable local oscillators, (3) custom CMOS gate-array LSTs of 1-b quadraphase correlators for 4/spl times/4 combinations, and (4) new image processing techniques to suppress large sidelobe effects due to the solar disk and extended sources.>