Pub Date : 2021-10-25DOI: 10.1109/SSRR53300.2021.9597693
J. Borer, M. Pryor
The proliferation of more affordable sensors technologies has enabled the development of a variety of localization modalities and fusion techniques for pose estimation. Utilizing multiple localization techniques provides a more accurate pose estimate as well as a more robust solution reducing the risk associated with a lack of data for any given sensor due to occlusion(s) which here refers to the unavailability of a sensor's data due to signal loss from physical attenuation, distance, low bandwidth, sensor limitations, or sensor failure. In large complex environments the impact of occlusions cannot be known a priori - temporally or spatially. Here, we present a novel heuristic-based GNSS carrier noise re-initialization framework to manage transitions between localization modalities. Disturbance rejection is used to eliminate discrete filter jitter and manage the competing interests of competing state estimate data sources. The hybrid localization method is evaluated in a relevant environment and shown to be more effective than each individual localization modality.
{"title":"Continuous Hybrid Localization in Environments with Physical and Temporal Sensor Occlusions","authors":"J. Borer, M. Pryor","doi":"10.1109/SSRR53300.2021.9597693","DOIUrl":"https://doi.org/10.1109/SSRR53300.2021.9597693","url":null,"abstract":"The proliferation of more affordable sensors technologies has enabled the development of a variety of localization modalities and fusion techniques for pose estimation. Utilizing multiple localization techniques provides a more accurate pose estimate as well as a more robust solution reducing the risk associated with a lack of data for any given sensor due to occlusion(s) which here refers to the unavailability of a sensor's data due to signal loss from physical attenuation, distance, low bandwidth, sensor limitations, or sensor failure. In large complex environments the impact of occlusions cannot be known a priori - temporally or spatially. Here, we present a novel heuristic-based GNSS carrier noise re-initialization framework to manage transitions between localization modalities. Disturbance rejection is used to eliminate discrete filter jitter and manage the competing interests of competing state estimate data sources. The hybrid localization method is evaluated in a relevant environment and shown to be more effective than each individual localization modality.","PeriodicalId":423263,"journal":{"name":"2021 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122227496","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-10-25DOI: 10.1109/SSRR53300.2021.9597854
Christopher M. Reardon, Kerstin S Haring, J. Gregory, J. Rogers
Online change detection performed by mobile robots has incredible potential to impact safety and security applications. While robots are superior to humans at detecting changes, humans are still better at interpreting this information and will be responsible for making critical decisions in these contexts. For these reasons, robot-to-human communication of change detection is a fundamental requirement for successful human-robot teams operating in such scenarios. In this work we seek to improve this communication, and present the results of a study that evaluates the interpretability of autonomous robot-based change detections conveyed via mixed reality to untrained human participants. Our results show that humans are able to identify changes and understand the visualizations employed without prior training. Our analysis of the limitations of this initial study should be constructive to future work in this domain.
{"title":"Evaluating Human Understanding of a Mixed Reality Interface for Autonomous Robot-Based Change Detection","authors":"Christopher M. Reardon, Kerstin S Haring, J. Gregory, J. Rogers","doi":"10.1109/SSRR53300.2021.9597854","DOIUrl":"https://doi.org/10.1109/SSRR53300.2021.9597854","url":null,"abstract":"Online change detection performed by mobile robots has incredible potential to impact safety and security applications. While robots are superior to humans at detecting changes, humans are still better at interpreting this information and will be responsible for making critical decisions in these contexts. For these reasons, robot-to-human communication of change detection is a fundamental requirement for successful human-robot teams operating in such scenarios. In this work we seek to improve this communication, and present the results of a study that evaluates the interpretability of autonomous robot-based change detections conveyed via mixed reality to untrained human participants. Our results show that humans are able to identify changes and understand the visualizations employed without prior training. Our analysis of the limitations of this initial study should be constructive to future work in this domain.","PeriodicalId":423263,"journal":{"name":"2021 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114083907","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-10-25DOI: 10.1109/SSRR53300.2021.9597857
Bálint Kövári, E. Ebeid
AI-based autonomous onboard drone applications are evolving rapidly and demand dedicated hardware resources to perform effectively. Currently, CPUs and GPUs are commonly used to run these applications. This paper presents a novel drone platform called MPDrone based on the cutting-edge MPSoC boards that combine FPGA, CPU, and GPU in a single chip. The proposed platform utilizes the reconfigurable FPGA chip to run heavy AI algorithms and the CPU to execute ROS for processing communication with the drone flight controller and onboard sensors. The paper introduces the design and implementation of the MPDrone platform, which is validated in simulation and real-world testing through an intelligent object detection and landing use case. The testing results proved the applicability of the proposed FPGA-based platform for AI applications.
{"title":"MPDrone: FPGA-based Platform for Intelligent Real-time Autonomous Drone Operations","authors":"Bálint Kövári, E. Ebeid","doi":"10.1109/SSRR53300.2021.9597857","DOIUrl":"https://doi.org/10.1109/SSRR53300.2021.9597857","url":null,"abstract":"AI-based autonomous onboard drone applications are evolving rapidly and demand dedicated hardware resources to perform effectively. Currently, CPUs and GPUs are commonly used to run these applications. This paper presents a novel drone platform called MPDrone based on the cutting-edge MPSoC boards that combine FPGA, CPU, and GPU in a single chip. The proposed platform utilizes the reconfigurable FPGA chip to run heavy AI algorithms and the CPU to execute ROS for processing communication with the drone flight controller and onboard sensors. The paper introduces the design and implementation of the MPDrone platform, which is validated in simulation and real-world testing through an intelligent object detection and landing use case. The testing results proved the applicability of the proposed FPGA-based platform for AI applications.","PeriodicalId":423263,"journal":{"name":"2021 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114139105","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-10-25DOI: 10.1109/SSRR53300.2021.9597856
Martin Pallin, Jayedur Rashid, Petter Ögren
In this paper we propose a version of the Genetic Algorithm (GA) for combined task assignment and path planning that is highly decentralized in the sense that each agent only knows its own capabilities and data, and a set of so-called handover values communicated to it from the other agents over an unreliable low bandwidth communication channel. These handover values are used in combination with a local GA involving no other agents, to decide what tasks to execute, and what tasks to leave to others. We compare the performance of our approach to a centralized version of GA, and a partly decentralized version of GA where computations are local, but all agents need complete information regarding all other agents, including position, range, battery, and local obstacle maps. We compare solution performance as well as messages sent for the three algorithms, and conclude that the proposed algorithms has a small decrease in performance, but a significant decrease in required communication. We compare the performance of our approach to a centralized version of GA, and a partly decentralized version of GA where computations are local, but all agents need complete information regarding all other agents, including position, range, battery, and local obstacle maps. We compare solution performance as well as messages sent for the three algorithms, and conclude that the proposed algorithms has a small decrease in performance, but a significant decrease in required communication. We compare solution performance as well as messages sent for the three algorithms, and conclude that the proposed algorithms has a small decrease in performance, but a significant decrease in required communication.
{"title":"A Decentralized Asynchronous Collaborative Genetic Algorithm for Heterogeneous Multi-agent Search and Rescue Problems","authors":"Martin Pallin, Jayedur Rashid, Petter Ögren","doi":"10.1109/SSRR53300.2021.9597856","DOIUrl":"https://doi.org/10.1109/SSRR53300.2021.9597856","url":null,"abstract":"In this paper we propose a version of the Genetic Algorithm (GA) for combined task assignment and path planning that is highly decentralized in the sense that each agent only knows its own capabilities and data, and a set of so-called handover values communicated to it from the other agents over an unreliable low bandwidth communication channel. These handover values are used in combination with a local GA involving no other agents, to decide what tasks to execute, and what tasks to leave to others. We compare the performance of our approach to a centralized version of GA, and a partly decentralized version of GA where computations are local, but all agents need complete information regarding all other agents, including position, range, battery, and local obstacle maps. We compare solution performance as well as messages sent for the three algorithms, and conclude that the proposed algorithms has a small decrease in performance, but a significant decrease in required communication. We compare the performance of our approach to a centralized version of GA, and a partly decentralized version of GA where computations are local, but all agents need complete information regarding all other agents, including position, range, battery, and local obstacle maps. We compare solution performance as well as messages sent for the three algorithms, and conclude that the proposed algorithms has a small decrease in performance, but a significant decrease in required communication. We compare solution performance as well as messages sent for the three algorithms, and conclude that the proposed algorithms has a small decrease in performance, but a significant decrease in required communication.","PeriodicalId":423263,"journal":{"name":"2021 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128105934","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-10-25DOI: 10.1109/SSRR53300.2021.9597678
Alana Sanchez, W. Smart
Robots are being increasingly used in the fight against highly-infectious diseases such as Ebola, MERS, and SARS-CoV-2. Many of these robots use ultraviolet lights mounted on a mobile base to inactivate the pathogens. While the lights are generally effective at irradiating open spaces and walls, they are less effective when it comes to horizontal surfaces, because of the orientation of the light sources. This can be problematic for pathogens such as Ebola, where transmission via contaminated work surfaces, which are often horizontal, is a concern. In this paper, we describe the design, implementation, and testing of an ultraviolet light disinfection system implemented on a mobile manipulator robot designed to address the problem of horizontal surface disinfection. A human supervisor designates a surface for disinfection, the robot autonomously plans and executes an end-effector trajectory to disinfect the surface to the required certainty, and then displays the results for the supervisor to verify. We also provide some background information on Ultraviolet Germicidal Irradiation (UVGI) and describe how we constructed and validated models of ultraviolet radiation propagation and accumulation in our system. Finally, we describe our implementation on a Fetch mobile manipulation platform, and discuss how the practicalities of implementation on a real robot affect our models.
{"title":"A Shared Autonomy Surface Disinfection System Using a Mobile Manipulator Robot","authors":"Alana Sanchez, W. Smart","doi":"10.1109/SSRR53300.2021.9597678","DOIUrl":"https://doi.org/10.1109/SSRR53300.2021.9597678","url":null,"abstract":"Robots are being increasingly used in the fight against highly-infectious diseases such as Ebola, MERS, and SARS-CoV-2. Many of these robots use ultraviolet lights mounted on a mobile base to inactivate the pathogens. While the lights are generally effective at irradiating open spaces and walls, they are less effective when it comes to horizontal surfaces, because of the orientation of the light sources. This can be problematic for pathogens such as Ebola, where transmission via contaminated work surfaces, which are often horizontal, is a concern. In this paper, we describe the design, implementation, and testing of an ultraviolet light disinfection system implemented on a mobile manipulator robot designed to address the problem of horizontal surface disinfection. A human supervisor designates a surface for disinfection, the robot autonomously plans and executes an end-effector trajectory to disinfect the surface to the required certainty, and then displays the results for the supervisor to verify. We also provide some background information on Ultraviolet Germicidal Irradiation (UVGI) and describe how we constructed and validated models of ultraviolet radiation propagation and accumulation in our system. Finally, we describe our implementation on a Fetch mobile manipulation platform, and discuss how the practicalities of implementation on a real robot affect our models.","PeriodicalId":423263,"journal":{"name":"2021 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115390512","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-10-25DOI: 10.1109/SSRR53300.2021.9597683
Amedeo Fabris, Steffen Kirchgeorg, S. Mintchev
In post-disaster scenarios, rescuers are often confronted with the challenge of accessing confined and cluttered environments including long and narrow passageways, gaps in walls or ceilings. Because of their mobility and versatility, there is a growing interest in developing drones for the remote exploration of these dangerous and often difficult to access places. However, the mechanical design and locomotion strategies of current drones limit the size of the confined space that can be explored. In this work, we present a quadcopter capable of traversing long passageways 34% smaller than its nominal size. The combination of a soft morphing frame and multi-modal mobility allows the drone to exploit a new dynamic strategy for passageway traversal. The drone flies at a given speed towards the entrance of the passageway until it collides with it. The momentum and ability of the frame to soften allow the drone to passively fold and enter. Once the drone is squeezed between the walls of the passageway, it uses two tracks to crawl through. Through experiments, we characterize the main mechanical systems of the drone and study the entry into crevices of different sizes.
{"title":"A Soft Drone with Multi-modal Mobility for the Exploration of Confined Spaces","authors":"Amedeo Fabris, Steffen Kirchgeorg, S. Mintchev","doi":"10.1109/SSRR53300.2021.9597683","DOIUrl":"https://doi.org/10.1109/SSRR53300.2021.9597683","url":null,"abstract":"In post-disaster scenarios, rescuers are often confronted with the challenge of accessing confined and cluttered environments including long and narrow passageways, gaps in walls or ceilings. Because of their mobility and versatility, there is a growing interest in developing drones for the remote exploration of these dangerous and often difficult to access places. However, the mechanical design and locomotion strategies of current drones limit the size of the confined space that can be explored. In this work, we present a quadcopter capable of traversing long passageways 34% smaller than its nominal size. The combination of a soft morphing frame and multi-modal mobility allows the drone to exploit a new dynamic strategy for passageway traversal. The drone flies at a given speed towards the entrance of the passageway until it collides with it. The momentum and ability of the frame to soften allow the drone to passively fold and enter. Once the drone is squeezed between the walls of the passageway, it uses two tracks to crawl through. Through experiments, we characterize the main mechanical systems of the drone and study the entry into crevices of different sizes.","PeriodicalId":423263,"journal":{"name":"2021 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR)","volume":"139 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115973012","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-10-25DOI: 10.1109/SSRR53300.2021.9597865
Bohan Yoon, Hyeonha Kim, Geonsik Youn, J. Rhee
Recently, drones have been used more in various fields such as safety, security, and rescue. Drones have the advantage of being able to explore in a wide range through the camera mounted on the drone. In the field of inventory management automation, research was conducted to utilize it. For inventory management automation in a large warehouse, a camera mounted on the drone scan pre-displayed ground QR (Quick Response) code to explore the path. The drone runs along the navigated path and manages the inventory of the warehouse by scanning the barcode or QR code attached to the product. However, unlike warehouses, which have well-defined grids or shelves, the location where products are stored in a yard is not fixed but flexible. Thus, for efficient inventory management in the storage yard, it is also necessary to estimate the position of the QR codes attached to the product. Therefore, in this paper, we propose a position estimation method for drones and products based on the QR code segmentation model. The segmentation model is used to detect the region of perspective distortion QR code caused by the angle difference between the camera and the QR code. Subsequently, shape correction and decoding of the detected QR code region are performed to determine whether it is a ground QR code or not, and the position of the drone is estimated. Finally, the 3D coordinates of the QR code attached to the product, not the ground QR code, are calculated from images taken by drones from two different viewpoints. Consequently, the 3D position coordinates of the drones and QR codes attached to the products will be estimated using the ground QR codes, and efficient inventory management in the storage yard will be achieved in this way.
{"title":"3D position estimation of drone and object based on QR code segmentation model for inventory management automation","authors":"Bohan Yoon, Hyeonha Kim, Geonsik Youn, J. Rhee","doi":"10.1109/SSRR53300.2021.9597865","DOIUrl":"https://doi.org/10.1109/SSRR53300.2021.9597865","url":null,"abstract":"Recently, drones have been used more in various fields such as safety, security, and rescue. Drones have the advantage of being able to explore in a wide range through the camera mounted on the drone. In the field of inventory management automation, research was conducted to utilize it. For inventory management automation in a large warehouse, a camera mounted on the drone scan pre-displayed ground QR (Quick Response) code to explore the path. The drone runs along the navigated path and manages the inventory of the warehouse by scanning the barcode or QR code attached to the product. However, unlike warehouses, which have well-defined grids or shelves, the location where products are stored in a yard is not fixed but flexible. Thus, for efficient inventory management in the storage yard, it is also necessary to estimate the position of the QR codes attached to the product. Therefore, in this paper, we propose a position estimation method for drones and products based on the QR code segmentation model. The segmentation model is used to detect the region of perspective distortion QR code caused by the angle difference between the camera and the QR code. Subsequently, shape correction and decoding of the detected QR code region are performed to determine whether it is a ground QR code or not, and the position of the drone is estimated. Finally, the 3D coordinates of the QR code attached to the product, not the ground QR code, are calculated from images taken by drones from two different viewpoints. Consequently, the 3D position coordinates of the drones and QR codes attached to the products will be estimated using the ground QR codes, and efficient inventory management in the storage yard will be achieved in this way.","PeriodicalId":423263,"journal":{"name":"2021 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR)","volume":"94 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121700529","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-10-25DOI: 10.1109/SSRR53300.2021.9597676
A. Pham, Anh T. La, Ethan Chang, Hung M. La
The research of robots to assist people in inspecting the quality of steel bridges has attracted significant attention in recent years. However, the intricate structure of the steel bridge components poses a massive challenge for researchers to move the robot across the bridge to perform the tests. This paper presents a new development of a hybrid flying-climbing robotic system, which can move flexibly and quickly to different positions on the steel bridge. In addition to using high-resolution cameras for an overview, the design allows the robot to stick to steel surfaces and act as a mobile robot for more detailed inspection with our developed giant magneto-resistance (GMR) sensor array system. We conduct a mechanical analysis to show the climbing capability of the mobile part. Additionally, we develop a landing algorithm to allow the robot to land on a steel surface to perform in-depth inspection safely. The designed GMR sensor array has shown the capability of detecting steel cracks to support the in-depth inspection mode. We have tested and validated our developed robot on real bridges to ensure that the design works well and is stable.
{"title":"Flying-Climbing Mobile Robot for Steel Bridge Inspection","authors":"A. Pham, Anh T. La, Ethan Chang, Hung M. La","doi":"10.1109/SSRR53300.2021.9597676","DOIUrl":"https://doi.org/10.1109/SSRR53300.2021.9597676","url":null,"abstract":"The research of robots to assist people in inspecting the quality of steel bridges has attracted significant attention in recent years. However, the intricate structure of the steel bridge components poses a massive challenge for researchers to move the robot across the bridge to perform the tests. This paper presents a new development of a hybrid flying-climbing robotic system, which can move flexibly and quickly to different positions on the steel bridge. In addition to using high-resolution cameras for an overview, the design allows the robot to stick to steel surfaces and act as a mobile robot for more detailed inspection with our developed giant magneto-resistance (GMR) sensor array system. We conduct a mechanical analysis to show the climbing capability of the mobile part. Additionally, we develop a landing algorithm to allow the robot to land on a steel surface to perform in-depth inspection safely. The designed GMR sensor array has shown the capability of detecting steel cracks to support the in-depth inspection mode. We have tested and validated our developed robot on real bridges to ensure that the design works well and is stable.","PeriodicalId":423263,"journal":{"name":"2021 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131218176","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-10-25DOI: 10.1109/SSRR53300.2021.9597866
Moritz Torchalla, Marius Schnaubelt, Kevin Daun, O. Stryk
We address the problem of robust simultaneous mapping and localization in degraded visual conditions using low-cost off-the-shelf radars. Current methods often use high-end radar sensors or are tightly coupled to specific sensors, limiting the applicability to new robots. In contrast, we present a sensor-agnostic processing pipeline based on a novel forward sensor model to achieve accurate updates of signed distance function-based maps and robust optimization techniques to reach robust and accurate pose estimates. Our evaluation demonstrates accurate mapping and pose estimation in indoor environments under poor visual conditions and higher accuracy compared to existing methods on publicly available benchmark data.
{"title":"Robust Multisensor Fusion for Reliable Mapping and Navigation in Degraded Visual Conditions","authors":"Moritz Torchalla, Marius Schnaubelt, Kevin Daun, O. Stryk","doi":"10.1109/SSRR53300.2021.9597866","DOIUrl":"https://doi.org/10.1109/SSRR53300.2021.9597866","url":null,"abstract":"We address the problem of robust simultaneous mapping and localization in degraded visual conditions using low-cost off-the-shelf radars. Current methods often use high-end radar sensors or are tightly coupled to specific sensors, limiting the applicability to new robots. In contrast, we present a sensor-agnostic processing pipeline based on a novel forward sensor model to achieve accurate updates of signed distance function-based maps and robust optimization techniques to reach robust and accurate pose estimates. Our evaluation demonstrates accurate mapping and pose estimation in indoor environments under poor visual conditions and higher accuracy compared to existing methods on publicly available benchmark data.","PeriodicalId":423263,"journal":{"name":"2021 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR)","volume":"142 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116573266","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-10-25DOI: 10.1109/SSRR53300.2021.9597677
H. Surmann, Dominik Slomma, Stefan Grobelny, R. Grafe
This technical report is about the mission and the experience gained during the reconnaissance of an industrial hall with hazardous substances after a major fire in Berlin. During this operation, only UAVs and cameras were used to obtain information about the site and the building. First, a geo-referenced 3D model of the building was created in order to plan the entry into the hall. Subsequently, the UAVs were used to fly in the heavily damaged interior and take pictures from inside of the hall. A 360° camera mounted under the UAV was used to collect images of the surrounding area especially from sections that were difficult to fly into. Since the collected data set contained similar images as well as blurred images, it was cleaned from non-optimal images using visual SLAM, bundle adjustment and blur detection so that a 3D model and overviews could be calculated. It was shown that the emergency services were not able to extract the necessary information from the 3D model. Therefore, an interactive panorama viewer with links to other 360° images was implemented where the links to the other images depends on the semi dense point cloud and located camera positions of the visual SLAM algorithm so that the emergency forces could view the surroundings.
{"title":"Deployment of Aerial Robots after a major fire of an industrial hall with hazardous substances, a report","authors":"H. Surmann, Dominik Slomma, Stefan Grobelny, R. Grafe","doi":"10.1109/SSRR53300.2021.9597677","DOIUrl":"https://doi.org/10.1109/SSRR53300.2021.9597677","url":null,"abstract":"This technical report is about the mission and the experience gained during the reconnaissance of an industrial hall with hazardous substances after a major fire in Berlin. During this operation, only UAVs and cameras were used to obtain information about the site and the building. First, a geo-referenced 3D model of the building was created in order to plan the entry into the hall. Subsequently, the UAVs were used to fly in the heavily damaged interior and take pictures from inside of the hall. A 360° camera mounted under the UAV was used to collect images of the surrounding area especially from sections that were difficult to fly into. Since the collected data set contained similar images as well as blurred images, it was cleaned from non-optimal images using visual SLAM, bundle adjustment and blur detection so that a 3D model and overviews could be calculated. It was shown that the emergency services were not able to extract the necessary information from the 3D model. Therefore, an interactive panorama viewer with links to other 360° images was implemented where the links to the other images depends on the semi dense point cloud and located camera positions of the visual SLAM algorithm so that the emergency forces could view the surroundings.","PeriodicalId":423263,"journal":{"name":"2021 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131051057","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：