Nobuhiro Shimoi

Development of a wheeled mobile robot "octal wheel" realized climbing up and down stairs

This paper proposes an eight-wheeled robot which is able to climb over the uneven terrain for rescue, de-mining other works. In order to perform these works without human assistance, robots must have the ability to move on rugged terrain. Wheeled vehicles have advantages for moving efficiency and speed but the disadvantage is that the diameter of the wheel limits which obstacles can be surmounted. This paper proposes a mechanism which eliminates the disadvantages of a wheeled system. This mechanism is applied to a self-standing type eight-wheeled robot which is able to climb up and down stairs by utilizing a command form remote controller. Experimental results demonstrate the effectiveness of this mechanism and robot.

Stable sparse channel estimation algorithm under non-Gaussian noise environments

Broadband frequency-selective fading channels usually exhibit the inherent sparse structure distribution in spread time-domain. By exploiting the sparsity, adaptive sparse channel estimation (ASCE) algorithms, e.g., least mean square with reweighted L1-norm constraint (LMS-RL1) algorithm, can bring a considerable performance gain under the assumption of additive white Gaussian noise (AWGN). In the scenarios of real wireless communication systems, however, channel estimation performance is often deteriorated by the unexpected non-Gaussian mixture noises which usually include AWGN and impulsive noises. To design stable communication systems, we propose sign LMS-RL1 (SLMS-RL1) channel estimation algorithm to remove the non-Gaussian noises and to exploit channel sparsity simultaneously. In addition, the regularization parameter (REPA) selection for SLMS-RL1 algorithm is proposed via Monte Carlo method. Simulation results are provided to corroborate our studies.

Remote mine sensing technology using a mobile wheeled robot RAT-1

To conduct mine detection experiments using an octal wheeled prototype robot RAT-1, we developed end effectors to be attached to the driving wheels of the robot. This enables the robot to step safely and stably without hitting hidden mines. We created a simulation model for this study to test the movement of a robot having metal sensors attached to the front of its wheels and a driving algorithm with effect control based on IR cameras. We verified the efficiency of the system in actual walking experiments. We also studied remote sensing technology uses for IR cameras combined with other metal sensors. Tests with trial mines were used to study the detection characteristics of IR cameras and various technologies for collecting and processing image data in real time for optimum mine detection.

Adaptive Category Mapping Networks for all-mode topological feature learning used for mobile robot vision

This paper presents an adaptive and incremental learning method to visualize series data on a category map. We designate this method as Adaptive Category Mapping Networks (ACMNs). The architecture of ACMNs comprises three modules: a codebook module, a labeling module, and a mapping module. The codebook module converts input features into codebooks as low-dimensional vectors using Self-Organizing Maps (SOMs). The labeling module creates labels as a candidate of categories based on the incremental learning of Adaptive Resonance Theory (ART). The mapping module visualizes spatial relations among categories on a category map using Counter Propagation Networks (CPNs). ACMNs actualize supervised, semi-supervised, and unsupervised learning as all-mode learning to switch network structures including connections. The experimentally obtained results obtained using two open datasets reveal that the recognition accuracy of our method is superior to that of the former method. Moreover, we address applications of the visualizing function using category maps.

Improved adaptive sparse channel estimation using re-weighted L1-norm normalized least mean fourth algorithm

In the frequency-selective fading broadband wireless communications systems, two adaptive sparse channel estimation (ASCE) methods using zero-attracting normalized least mean fourth (ZA-NLMF) algorithm and reweighted ZA-NLMF (RZA-NLMF) algorithm have been proposed to mitigate noise and to exploit channel sparsity. Motivated by compressive sensing, in this paper, an improved ASCE method is proposed by using reweighted L1-norm NLMF (RL1-NLMF) algorithm where RL1 can exploit more sparsity information than ZA and RZA. Specifically, we construct the cost function of RL1-NLMF algorithm and hereafter derive its update equation. Intuitive illustration is also given to demonstrate that RL1 is more efficient than conventional two sparsity constraints. Finally, simulation results are provided to show that the proposed method achieves better estimation performance than the two conventional ones.

Estimation of dynamic properties of traditional wooden structures using new bolt sensor

Dynamic properties of traditional wooden structures are estimated from forced vibration test performed on a prototype constructed for this purpose. The prototype corresponds to a framed wooden construction with traditional connections between columns and beams without nails and with wedges inserted into joints to fix them. To verify the applicability of a new type of piezoelectric bolt sensor the test series were performed using also commercial accelerometers and laser displacement transducer for comparison. The new bolt sensor is intended to be used for structural health monitoring of important and small structures like historical shrines or other small historical buildings. Bolt sensors were installed in selected frame joints and changes in the voltage signal were detected when the prototype is subjected to dynamic excitation. The response of the new sensor is comparable with that obtained by high precision commercial accelerometers and laser displacement transducer. In addition the dynamic response of the structure and the response of the bolt senor were verified analytically using finite element method. For analytical modeling semi-rigid joint is used where the moment rotation relationship is specified for each beam end. The research serves also to calibrate the analytical model by using experimental results obtained from forced vibration test.

Implementing an In-Home Sensor Agent in Conjunction with an Elderly Monitoring Network

In this paper, we present the design and implementation of the in-home sensor agent MaMoRu-Kun as an Internet of Things (IoT) smart device, developed via the “Research and development of the regional/solitary elderly life support system using multi-fusion sensors” project. At Akita Prefectural University, the bed and pillow sensors and corresponding monitoring system have been developed to watch elderly individuals at bedtime, in particular those who live alone. As a sensor agent, MaMoRu-Kun is connected to the in-home wireless network of the target individuals accommodations and collects trigger information from various switches, motion detection sensors, and a remote controller. This smart device is also able to send its entire set of data along with the status of the sensor to a collection and monitoring server connected via a long-term evolution (LTE) router. We implemented this agent using an Arduino and a Bluetooth-connected Android terminal.

Context based semantic scene classification and recognition used for a vision-based mobile robot

This paper presents a novel method using accelerated KAZE (AKAZE) and Gist for a context-based semantic classification and recognition of indoor scenes used for a vision-based mobile robot. Our method represents spatial relations among categories for mapping neighborhood units on category maps using counter propagation networks (CPNs) while maintaining sequential information of labels generated from adaptive resonance theory 2 (ART-2) networks. We evaluated the performance and accuracy of semantic categories using KTH-IDOL benchmark datasets. Compared with the earlier described method using scale-invariant feature transform (SIFT), accuracies of classification, recognition, and F-measure were improved to 2.9%, 3.4%, and 3.3% of our method using AKAZE. For analyzing results, confusion matrixes show that incorrect images between corridors and rooms are decreased in our method compared with the former method. We consider that our proposed feature representation method based on context and category formation, combined with ART-2 and CPNs, is useful for indoor scene classification and recognition for robot vision.

Adaptive learning based driving episode description on category maps

This study was conducted to create driving episodes using machine-learning-based algorithms that address long-term memory (LTM) and topological mapping. This paper presents a novel episodic memory model for driving safety according to traffic scenes. The model incorporates three important features: adaptive resonance theory (ART), which learns time-series features incrementally while maintaining stability and plasticity for time-series data; self-organizing maps (SOMs), which represent input data as a map with topological relations using self-mapping characteristics; and counter propagation networks (CPNs), which label category maps using input features and counter signals. Category maps represent driving episode information that includes driving contexts and facial expressions. The bursting states of respective maps produce LTM, which is created on ART as episodic memory. Evaluation of the experimentally obtained results show the possibility of using recorded driving episodes with image datasets obtained using an event data recorder (EDR) with two cameras. Using category maps, we visualize driving features according to driving scenes on a public road and an expressway.

A Multi-agent Approach to Smart Home Sensors for the Elderly based on an Open Hardware Architecture: A Model for Participatory Evaluation.

In this position paper, we present the design and implementation of an in-home sensor agent as an Internet of Things (IoT) smart device system based on an open hardware architecture. This sensor agent is designed to be connected to a wireless network in the target individual’s home where it collects trigger information from various switches, motion detection sensors in the room, along with the pillow and bed. We began our study by conducting a complete requirement analysis to determine the functions required for the home sensors. Next, we examined the proposed system terms of its hardware and software requirements and fabricated working prototypes. Here, it should be noted that the hardware and software were designed to aggregate connections with various composite sensor devices in order to allow trigger collection and processing. Finally, we reviewed visualization techniques for displaying the analytical results of the monitoring under normal conditions and emergency situations.