This letter presents benchmark tests of various visual(-inertial) odometry algorithms on NVIDIA Jetson platforms. The compared include mono and stereo, covering Visual Odometry (VO) Visual-Inertial (VIO): VINS-Mono, VINS-Fusion, Kimera, ALVIO, Stereo-MSCKF, ORB-SLAM2 ROVIO. As these methods are mainly used for unmanned aerial vehicles (UAVs), they must perform well in situations where the size processing board weight is limited. boards released by satisfy constraints as have a sufficiently...

Perception of traversable regions and objects interest from a 3D point cloud is one the critical tasks in autonomous navigation. A ground vehicle needs to look for terrains that are explorable by wheels. Then, make safe navigation decisions, segmentation positioned on those has be followed up. However, over-segmentation under-segmentation can negatively influence such decisions. To end, we propose <italic xmlns:mml="http://www.w3.org/1998/Math/MathML"...

Wepropose a novel state estimator for legged robots, <i>STEP</i>, achieved through preintegrated foot velocity factor. In the factor, usual non-slip assumption is not adopted. Instead, end effector becomes observable by exploiting body speed obtained from stereo camera. other words, effector&#x2019;s pose can be estimated. Another advantage of our approach that it eliminates necessity contact detection step, unlike typical approaches. The proposed method has also been validated in...

Exploring an unknown environment without colliding with obstacles is one of the essentials autonomous vehicles to perform diverse missions such as structural inspections, rescues, deliveries, and so forth. Therefore, unmanned aerial (UAVS), which are fast, agile, have high degrees freedom, been widely used. However, previous approaches two limitations: a) First, they may not be appropriate for exploring large-scale environments because mainly depend on random sampling-based path planning...

Global registration is a fundamental task that estimates the relative pose between two viewpoints of 3D point clouds. However, there are issues degrade performance global in LiDAR SLAM: one sparsity issue and other degeneracy. The caused by sparse characteristics cloud measurements mechanically spinning sensor. degeneracy sometimes occurs because outlier-rejection methods reject too many correspondences, leaving less than three inliers. These have become more severe as discrepancy clouds...

For many years, there has been an impressive progress on visual odometry applied to mobile robots and drones. However, the perception is still in spotlight as a challenging field because vision sensor some problems obtaining correct scale information with monocular camera also vulnerable situation which illumination changed. In this paper, UWB fusion proposed inertial algorithm solution mitigate problem. We designed cost function based mutual considering UWB. Considering characteristic of...

This paper presents an online coverage path planner for mapping the surface of three-dimensional structures. Recent works on structural rely random walks, volumetric cellular decomposition, and viewpoint optimization. Unfortunately, approaches that utilize decomposition or optimization cannot perform their work due to heavy computational complexity, planners based walks show inefficient spatial visiting, which leads failure coverage. In this paper, a novel non-randomized real-time utilizes...

Studies that broaden drone applications into complex tasks require a stable control framework. Recently, deep reinforcement learning (RL) algorithms have been exploited in many studies for robot to accomplish tasks. Unfortunately, RL might not be suitable being deployed directly real-world platform due the difficulty interpreting learned policy and lack of stability guarantee, especially task such as wall-climbing drone. This letter proposes novel hybrid architecture reinforces nominal...

This paper presents a state estimator design method using multi-rate sampled-data for autonomous vehicle driving system. The proposed is designed by an affine matched T-S fuzzy model with the sampling information of each sensors. For tracking control system fusion sensors, overall structure error dynamic based estimator. verified experimental results on Husky unmanned ground (UGV) equipped light detection and ranging (LiDAR), camera encoder.

Recently, application of Unmanned Aerial Vehicle (UAV) attract much attention not only academic but also industrial community. To control and make the most UAV, state estimation is mandatory it should work accurately on Real-time computational-limited onboard PC. Especially, for cases that GNSS-disabled environments or indoor situation, Visual Inertial Odometry using camera IMU sensor has been actively studied. Generally, to research develop navigation algorithm, high-performanced...

Exploring an unknown environment with unmanned robots has attracted many interests in the robotics field to replace human resources. Especially, quadruped robots, which can operate challenging terrains and have high payload capacity, been widely researched. However, are facing contact constraints of legs, namely traversability. In this paper, explore environments safely, we propose a novel local exploration planner that utilizes precomputed trajectories check traversability collision at same...

This study deals with the problem of visual loop closure detection that occurs when there are limited camera field views. The proposed method generates a panoramic image using multiple images and creates sub-panorama through sliding window process. Then, learning-based descriptor is formed for robust feature even in illumination changes challenging environments. After that, index descriptor, which closest to query found, final candidate found has same raw database. Histogram equalization...

Thanks to the high payload capacity and mobility, quadruped robots have been intensively researched perform various tasks such as explorations, industrial mappings, rescues. However, little attention has paid finding traversable regions performing simultaneously. In this paper, target tracking safely, we propose a novel system that strictly plans traversable, obstacle-free trajectory while estimating robot state with state-of-the-art SLAM (simultaneously localization mapping) position neural...

Recently, Unmanned Aerial Vehicles (UAVs) have been widely researched to accomplish tasks such as structural inspection, drone racing, and exploration of an unknown environment. To make the most UAV correctly estimate its pose with inexpensive light-weight sensors Inertial Measurement Unit (IMU) cameras, Visual Odometry (VIO) has intensively researched. There were a few analyses on performance VIO methods where only estimated error utilization computing resources considered. However, them...

Unmanned Aerial Vehicles (UAVs) have been intensively used in various fields thanks to their excellent maneuverability. However, the short operation time of UAV is limiting further utilization UAV. The limitation can be overcome by landing on ground vehicles and charging battery them. Therefore, diverse researches done robust platforms. Unfortunately, existing autonomous methods such limitations that a) special tag or marker should attached site; b) visibility site must secured; c) platform...

Unmanned aerial vehicles (UAVs) have been widely used in complex applications, such as military, exploration, and rescue. Although there are many quadcopter the bi-copter like a coaxial helicopter has apparent energy efficiency scalability advantages. What makes challenging to use is difficulty control because additional mechanical structures essential for stable movement. This paper tackles this problem by proposing novel bi-rotor design called M-BRIC with rotatable weight rods...

Although a UAV itself is adaptable to variety of tasks, its low payload and short flight time limit usage. As way overcome the limitations UAV, multi-UAV system being researched. In order utilize Multi-UAV system, it essential estimate relative position each UAV. this paper, estimation method proposed that requires only range measurements from pair combinations. The positions UAVs are estimated by minimizing difference between distance values which calculated UAVs. performance verified five...

This paper presents benchmark tests of various visual(-inertial) odometry algorithms on NVIDIA Jetson platforms. The compared include mono and stereo, covering Visual Odometry (VO) Visual-Inertial (VIO): VINS-Mono, VINS-Fusion, Kimera, ALVIO, Stereo-MSCKF, ORB-SLAM2 ROVIO. As these methods are mainly used for unmanned aerial vehicles (UAVs), they must perform well in situations where the size processing board weight is limited. boards released by satisfy constraints as have a sufficiently...

Abstract In the previous study, we identified cancer-associated immune checkpoint 5 (CAIC5) as a new co-inhibitory molecule belonging to B7-related family. Typically, CAIC5 is expressed in tissue macrophages including Kupffer cells and peritoneal macrophages. We also found out human fusion protein greatly suppressed proliferation cytokine production T NKT respective stimulation assay. Since expression was tumor microenvironment, examined impact of cell-associated on cell function....

Abstract Despite progress in the treatment of multiple myeloma (MM), this disease still remains fatal majority patients. Thus, new immunotherapeutic approaches are required to eradicate minimal residual cells. Tumor cells can be a source tumor-specific antigens for immunotherapy. However, tumor have been weak promote immune responses and maintain with large amounts burden vivo. In study, we evaluated compared efficacy vaccination autologous loaded artificial ligand T cell receptor NKT cells,...

In recent years, exploration methods have been extensively researched for the purpose of mapping unknown environments with autonomous robots, aiming to replace humans. However, existing primarily focus on indoor environments, which may not be sufficient fast and efficient outdoor environments. this paper, we propose a novel path planner, THE-Planner, utilizes topological hierarchical approach map quickly efficiently. THE-Planner builds local global graphs. It generates graph nodes robot's...