Autonomous systems are becoming inherently ubiquitous with the advancements of computing and communication solutions enabling low-latency offloading real-time collaboration distributed devices. Decentralized technologies blockchain ledger (DLTs) playing a key role. At same time, advances in deep learning (DL) have significantly raised degree autonomy level intelligence robotic autonomous systems. While these technological revolutions were taking place, raising concerns terms data security...

Light detection and ranging (LiDAR) sensor has become one of the primary sensors in robotics autonomous system for high-accuracy situational awareness. In recent years, multi-modal LiDAR systems emerged, among them, LiDAR-as-a-camera provide not only 3D point clouds but also fixed-resolution 360°panoramic images by encoding either depth, reflectivity, or near-infrared light image pixels. This potentially brings computer vision capabilities on top potential itself. this paper, we are...

LiDAR-based simultaneous localization and mapping (SLAM) approaches have obtained considerable success in autonomous robotic systems. This is part owing to the high accuracy of robust SLAM algorithms emergence new lower-cost LiDAR products. study benchmarks current state-of-the-art with a multi-modal sensor setup, showcasing diverse scanning modalities (spinning solid state) sensing technologies, cameras, mounted on mobile computing platform. We extend our previous multi-LiDAR dataset...

Unmanned aerial vehicles (UAVs) are becoming largely ubiquitous with an increasing demand for data. Accurate navigation and localization, required precise data collection in many industrial applications, often relies on RTK GNSS. These systems, able of centimeter-level accuracy, require a setup calibration process relatively expensive. This paper addresses the problem accurate positioning UAVs through cooperative localization. Inexpensive ultra-wideband (UWB) transceivers installed both UAV...

Over the last decade, robotic perception algorithms have significantly benefited from rapid advances in deep learning (DL). Indeed, a significant amount of autonomy stack different commercial and research platforms relies on DL for situational awareness, especially vision sensors. This work explored potential general-purpose algorithms, specifically detection segmentation neural networks, processing image-like outputs advanced lidar Rather than three-dimensional point cloud data, this is, to...

The role of deep learning (DL) in robotics has significantly deepened over the last decade. Intelligent robotic systems today are highly connected that rely on DL for a variety perception, control and other tasks. At same time, autonomous robots being increasingly deployed as part fleets, with collaboration among becoming more relevant factor. From perspective collaborative learning, federated (FL) enables continuous training models distributed, privacy-preserving way. This paper focuses...

Lidar technology has evolved significantly over the last decade, with higher resolution, better accuracy, and lower cost devices available today. In addition, new scanning modalities novel sensor technologies have emerged in recent years. Public datasets enabled benchmarking of algorithms set standards for cutting edge technology. However, existing are not representative technological landscape, only a reduced number lidars available. This inherently limits development comparison...

Abstract Purpose of Review: Distributed ledger technologies (DLTs), particularly blockchain, are paving the way to securing and managing distributed large-scale systems autonomous agents. We look into how these moving out lab real world within robotics field. Recent Findings: Despite scalability real-world applicability concerns, new solutions have emerged that show resilience intermittent connectivity, as well scalable for managed or permissioned networks. Summary: present a review on...

Abstract The increased data transmission and number of devices involved in communications among distributed systems make it challenging yet significantly necessary to have an efficient reliable networking middleware. In robotics autonomous systems, the wide application ROS 2 brings possibility utilizing various middlewares together with DDS for better communication edge or between cloud. However, there is a lack comprehensive performance comparison integrating these 2. this study, we provide...

The increased data transmission and number of devices involved in communications among distributed systems make it challenging yet significantly necessary to have an efficient reliable networking middleware. In robotics autonomous systems, the wide application ROS\,2 brings possibility utilizing various middlewares together with DDS for better communication edge or between cloud. However, there is a lack comprehensive performance comparison integrating these ROS\,2. this study, we provide...

Keypoint detection and description play a pivotal role in various robotics autonomous applications, including Visual Odometry (VO), visual navigation, Simultaneous Localization And Mapping (SLAM). While myriad of keypoint detectors descriptors have been extensively studied conventional camera images, the effectiveness these techniques context LiDAR-generated i.e., reflectivity ranges has not assessed. These images gained attention due to their resilience adverse conditions, such as rain or...

Deep learning methods have revolutionized mobile robotics, from advanced perception models for an enhanced situational awareness to novel control approaches through re-inforcement learning. This paper explores the potential of federated distributed systems robots enabling collaboration on Internet Robotic Things. To demonstrate effectiveness such approach, we deploy wheeled in different indoor environments. We analyze performance a approach and compare it traditional centralized training...

Micro-aerial vehicles (MAVs) are becoming ubiquitous across multiple industries and application domains. Lightweight MAVs with only an onboard flight controller a minimal sensor suite (e.g., IMU, vision, vertical ranging sensors) have potential as mobile easily deployable sensing platforms. When deployed from ground robot, key parameter is relative localization between the robot MAV. This paper proposes novel method for tracking in lidar point clouds. In clouds, we consider speed distance of...

With the increasing prevalence of drones in various industries, navigation and tracking unmanned aerial vehicles (UAVs) challenging environments, particularly GNSS-denied areas, have become crucial concerns. To address this need, we present a novel multi-LiDAR dataset specifically designed for UAV tracking. Our includes data from spinning LiDAR, two solid-state LiDARs with different Field View (FoV) scan patterns, an RGB-D camera. This diverse sensor suite allows research on new challenges...

With the increasing use of drones across various industries, navigation and tracking these unmanned aerial vehicles (UAVs) in challenging environments, namely GNSS-denied have become critical issues. In this paper, we propose a novel method for ground-based UAV system using solid-state LiDAR, which dynamically adjusts LiDAR frame integration time based on distance to its speed. Our fuses two simultaneous scan frequencies high accuracy persistent tracking, enabling reliable state estimation...

Objective: To explore the therapeutic effect of virtual reality technology on speech function Broca Aphasia patients after stroke. Method: Eighteen with stroke were enrolled in Rehabilitation Medicine Department, Third Affiliated Hospital Sun Yat-sen University from December 2016 to August 2017. The divided into observation group and control by random number table. Both groups trained function. received regular training for 20 minutes per day, Virtual Reality (VR) day; was given conventional...

Ultra-wideband (UWB) technology is a mature that contested other wireless technologies in the advent of IoT but did not achieve same levels widespread adoption. In recent years, however, with its potential as ranging and localization solution, it has regained momentum. Within robotics field, UWB positioning systems are being increasingly adopted for localizing autonomous ground or aerial robots. Industrial (IIoT) domain, ad-hoc networking simultaneous also explored. This survey overviews...

Over the last decade, robotic perception algorithms have significantly benefited from rapid advances in deep learning (DL). Indeed, a significant amount of autonomy stack different commercial and research platforms relies on DL for situational awareness, especially vision sensors. This work explores potential general-purpose algorithms, specifically detection segmentation neural networks, processing image-like outputs advanced lidar Rather than three-dimensional point cloud data, this is, to...

Integrating multiple LiDAR sensors can significantly enhance a robot's perception of the environment, enabling it to capture adequate measurements for simultaneous localization and mapping (SLAM). Indeed, solid-state LiDARs bring in high resolution at low cost traditional spinning robotic applications. However, their reduced field view (FoV) limits performance, particularly indoors. In this paper, we propose tightly-coupled multi-modal multi-LiDAR-inertial SLAM system surveying tasks. By...

This letter presents a cooperative relative multi-robot localization design and experimental study. We propose flexible Monte Carlo approach leveraging particle filter to estimate states. The estimation can be based on inter-robot Ultra-Wideband (UWB) ranging onboard odometry alone or dynamically integrated with spatial object detections from stereo cameras mounted each robot. main contributions of this work are as follows. First, we show that single UWB range is enough the accurate states...