Learning geometric and photometric features from panoramic LiDAR scans for outdoor place categorization

Semantic place categorization, which is one of the essential tasks for autonomous robots and vehicles, allows them to have capabilities of self-decision and navigation in unfamiliar environments. In particular, outdoor places are more difficult targets than indoor ones due to perceptual variations, such as dynamic illuminance over 24 hours and occlusions by cars and pedestrians. This paper presents a novel method of categorizing outdoor places using convolutional neural networks (CNNs), which take omnidirectional depth/reflectance images obtained by 3D LiDARs as the inputs. First, we construct a large-scale outdoor place dataset named Multi-modal Panoramic 3D Outdoor (MPO) comprising two types of point clouds captured by two different LiDARs. They are labeled with six outdoor place categories: coast, forest, indoor/outdoor parking, residential area, and urban area. Second, we provide CNNs for LiDAR-based outdoor place categorization and evaluate our approach with the MPO dataset. Our results on the MPO dataset outperform traditional approaches and show the effectiveness in which we use both depth and reflectance modalities. To analyze our trained deep networks, we visualize the learned features.     

The effects of post-annealing on the performance of ZnO thin film transistors

In this study, we investigated the effects of a post-annealing process on the performance and stability of zinc oxide thin film transistors fabricated by atomic layer deposition. After the post-annealing process in ambient air at 250 °C for 2 h, the value of the saturation mobility increased from 1.2 to 1.8 cm/Vs, the subthreshold swing decreased from 0.53 to 0.34 V/decade, and the I_on/I_off ratio increased from 3.1 × 10⁶ to 1.7 × 10⁷. The positive bias stability was also enhanced after post-annealing. These results are related to the formation of another phase in which the difference of enthalpy between the semiconductor material and contact metal electrode causes the carrier concentration at the metal/semiconductor interface to increase, leading to decreased contact resistivity. Additionally, internal modification of the semiconductor/dielectric interface and/or improving the semiconductor structure, which is related to a change in the oxidation state of Zn through the incorporation of oxygen and/or hydroxide, can result in improved device performance.     

A lightweight stateful address autoconfiguration for 6LoWPAN

Sensor networks have become increasingly important in various areas, and most current applications require connectivity between sensor networks and the Internet. By being seamlessly integrated into IP network infrastructure, sensor network applications would benefit from standardized and established technology, as well as from the plethora of readily available applications. Preparing sensor networks for IP communication and integrating them into the IP network, however, present new challenges on the architecture and its functional blocks, e.g., the adaptation of the respective link technology for IP support, development of security mechanisms, and autoconfiguration to support ad hoc deployment. In this paper, we focus on the IPv6 address autoconfiguration issue and propose a proxy-based autoconfiguration protocol. The proposed protocol guarantees the assignment of a unique address to each node in the network. The protocol is simulated and implemented on off-the-shelf sensor network platforms. The experiment results show that our mechanism outperforms similar network address configuring mechanisms in terms of latency and overhead.     

Empirical evaluation of receiver‐based TCP delay control in CDMA2000 networks

Wide‐area broadband wireless technologies such as CDMA2000 often suffer from variable transfer rate and long latency. In particular, TCP window‐based rate control causes excessive buffering at the base station because of the lower transfer rate of the wireless link than that of the wired backhaul link. This performance characteristic of TCP further increases the end‐to‐end delay, and additional resources are required at the base station. This paper presents a practical mechanism to control the end‐to‐end TCP delay for CDMA2000 networks (or other similar wireless technologies). The key idea is to reduce and stabilize RTT (round‐trip time) by dynamically controlling the TCP advertised window size, based on a runtime measurement of the wireless channel condition at the mobile station. The proposed system has been implemented by modifying the Linux protocol stack. The experiment results, conducted on a commercial CDMA2000 1x network, show that the proposed scheme greatly reduces the TCP delay in non‐congested networks, while not sacrificing the TCP throughput in congested networks. Copyright © 2006 John Wiley & Sons, Ltd.     

Nanometer-scale oxide thin film transistor with potential for high-density image sensor applications

The integration of electronically active oxide components onto silicon circuits represents an innovative approach to improving the functionality of novel devices. Like most semiconductor devices, complementary-metal-oxide-semiconductor image sensors (CISs) have physical limitations when progressively scaled down to extremely small dimensions. In this paper, we propose a novel hybrid CIS architecture that is based on the combination of nanometer-scale amorphous In-Ga-Zn-O (a-IGZO) thin-film transistors (TFTs) and a conventional Si photo diode (PD). With this approach, we aim to overcome the loss of quantum efficiency and image quality due to the continuous miniaturization of PDs. Specifically, the a-IGZO TFT with 180 nm gate length is probed to exhibit remarkable performance including low 1/f noise and high output gain, despite fabrication temperatures as low as 200 °C. In particular, excellent device performance is achieved using a double-layer gate dielectric (Al?O?/SiO?) combined with a trapezoidal active region formed by a tailored etching process. A self-aligned top gate structure is adopted to ensure low parasitic capacitance. Lastly, three-dimensional (3D) process simulation tools are employed to optimize the four-pixel CIS structure. The results demonstrate how our stacked hybrid device could be the starting point for new device strategies in image sensor architectures. Furthermore, we expect the proposed approach to be applicable to a wide range of micro- and nanoelectronic devices and systems.     

Temperature-dependent performance of the polymer electrolyte membrane fuel cell using short-side-chain perfluorosulfonic acid ionomer

We report on polymer electrolyte membrane fuel cells (PEMFCs) that function at high temperature and low humidity conditions based on short-side-chain perfluorosulfonic acid ionomer (SSC-PFSA). The PEMFCs fabricated with both SSC-PFSA membrane and ionomer exhibit higher performances than those with long-side-chain (LSC) PFSA at temperatures higher than 100 °C. The SSC-PFSA cell delivers 2.43 times higher current density (0.524 A cm⁻¹) at a potential of 0.6 V than LSC-PFSA cell at 140 °C and 20% relative humidity (RH). Such a higher performance at the elevated temperature is confirmed from the better membrane properties that are effective for an operation of high temperature fuel cell. From the characterization technique of TGA, XRD, FT-IR, water uptake and tensile test, we found that the SSC-PFSA membrane shows thermal stability by higher crystallinity, and chemical/mechanical stability than the LSC-PFSA membrane at high temperature. These fine properties are found to be the factor for applying Aquivion™ E87-05S membrane rather than Nafion^® 212 membrane for a high temperature fuel cell.     

Proxy‐based failure detection in multimedia streaming environments

This paper presents an effective failure‐detection mechanism called the ‘dynamic threshold,’ which has specifically developed for use with multimedia streaming applications. Compared to the conventional failure‐detection methods, which normally use a fixed threshold to detect a server failure, the proposed mechanism detects the failure in an effective and timely manner by using a dynamically adjusted timeout value upon which queries are sent to suspected streaming servers running in a dynamic network environment. The failover mechanism based on the dynamic threshold was implemented in a real streaming environment. The empirical results show that the dynamic threshold mechanism performs better than the conventional method in terms of timeliness, with comparable accuracy. Copyright © 2006 John Wiley & Sons, Ltd.     

Experimental Analysis of Timing Validation Methods for Distributed Real-Time Systems

Scheduling jobs dynamically on processors is likely to achieve better performance in multiprocessor and distributed real-time systems. Exhaustive methods for determining whether all jobs complete by their deadlines, in systems that use modern priority-driven scheduling strategies, are often infeasible or unreliable since the execution time of each job may vary. We previously published research results on finding worst-case bounds and efficient algorithms for validating systems in which independent jobs have arbitrary release times and deadlines, and are scheduled on processors dynamically in a priority-driven manner. An efficient method has been proposed to determine how late the completion times of jobs can be in dynamic systems where the jobs are preemptable and nonmigratable. This paper further presents the performance characteristics of the proposed methods, and shows its soundness by providing extensive simulation results. The worst-case completion times of jobs obtained with the proposed methods are compared with the values by simulations under different workload characteristics. The simulation results show that the proposed algorithm performs considerably well for diverse workloads. Considering the previous work showed the unlikelihood of finding tighter bounds than the one given in the paper, the simulation results indicate that the proposed methods effectively constitute a theoretical basis needed for a comprehensive validation strategy that is capable of dealing with dynamic distributed real-time systems.     

Categorization of indoor places by combining local binary pattern histograms of range and reflectance data from laser range finders

This paper presents an approach to categorize typical places in indoor environments using 3D scans provided by a laser range finder. Examples of such places are offices, laboratories, or kitchens. In our method, we combine the range and reflectance data from the laser scan for the final categorization of places. Range and reflectance images are transformed into histograms of local binary patterns and combined into a single feature vector. This vector is later classified using support vector machines. The results of the presented experiments demonstrate the capability of our technique to categorize indoor places with high accuracy. We also show that the combination of range and reflectance information improves the final categorization results in comparison with a single modality.     

Improving energy efficiency for flash memory based embedded applications

The JFFS2 file system for flash memory compresses files before actually writing them into flash memory. Because of this, multimedia files, for instance, which are already compressed in the application level go through an unnecessary and time-consuming compression stage and cause energy waste. Also, when reading such multimedia files, the default use of disk cache results in unnecessary main memory access, hence an energy waste, due to the low cache hit ratio. This paper presents two techniques to reduce the energy consumption of the JFFS2 flash file system for power-aware applications. One is to avoid data compression selectively when writing files, and the other is to bypass the page caching when reading sequential files. The modified file system is implemented on a PDA running Linux and the experiment results show that the proposed mechanism effectively reduces the overall energy consumption when accessing continuous and large files.