Vision-Based Autonomous UAV Navigation and Landing for Urban Search and Rescue

Mittal, Mayank; Mohan, Renjith; Burgard, Wolfram; Valada, Abhinav

doi:10.1007/978-3-030-95459-8_35

Cited by 22 publications

(15 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While they demonstrate it is possible to reach a human-like level of performance in real-world experiments by relying on RGB images only, their approach cannot be utilized for emergency landing in unknown environments, since it relies on the presence of a marker and does not deal with obstacles and occlusions. To overcome this limitation, more sophisticated vision-based pipelines that do not rely on fiducial markers have been proposed, using either geometrical [3], [4], [8] or deep learning approaches [7], [13]. These solutions are more flexible and allow robots to land safely in unstructured and unknown environments [8].…”

Section: Related Workmentioning

confidence: 99%

“…The work by Foster et al [4] explicitly creates an elevation map using depth completion to identify a suitable landing spot, and then generates a path to that position in a separate path-planning module. Similarly, Mittal et al [3] identify a safe landing area by processing dense depth images from a stereo camera and extracting terrain information, such as steepness and flatness, while also considering depth accuracy and the estimated energy consumption to reach a candidate location. Once a final position is identified, a safe path to the landing area is computed.…”

Section: Related Workmentioning

confidence: 99%

“…To run onboard drones with limited computing capabilities, the proposed pipeline is designed in a modular fashion, that allows to outsource depth estimation and semantic segmentation to cloud-based resources, if needed. We compare our approach against both a stateof-the-art landing pipeline [3] and a commercially available The input to the system is a monocular RGB image, which is used by a semantic segmentation NN [18] to generate a semantic mask of the environment. A depth completion network [17] outputs a dense depth map of the surroundings using the RGB image and sparse depth information.…”

Section: Related Workmentioning

confidence: 99%

“…of multi-rotor drones, such UAVs are generally equipped with a minimal sensor set-up composed of cameras and range sensors, rendering the emergency landing problem even more challenging, as a suitable landing location has to be identified by relying on limited sensory information. Recent solutions proposed in the literature either require manual intervention at the landing area by placing fiducial markers to identify a safe spot on the ground [1], [2], or require dense online 3D mapping of the environment [3], [4]. While the former methods are not suitable for emergency landing in unstructured environments, the latter ones, despite being more flexible, require high-quality sensory inputs.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Autonomous Emergency Landing for Multicopters using Deep Reinforcement Learning

Bartolomei

Kompis

Teixeira

et al. 2022

2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

View full text Add to dashboard Cite

This work presents a pipeline for autonomous emergency landing for multicopters, such as rotary wing Unmanned Aerial Vehicles (UAVs), using deep Reinforcement Learning (RL). Mechanical malfunctions, strong winds, sudden battery life drops (e.g. due to cold weather), failure in localization or GPS jamming are not uncommon and all constitute emergency situations that require a UAV to abort its mission early and land as quickly as possible in the immediate vicinity. To this end, it is crucial for a UAV that is deployed in real missions to be able to detect a safe landing spot efficiently and proceed to land autonomously, avoiding damage to both its integrity and the surroundings. Driven by the advances in semantic segmentation and depth completion using machine learning, the proposed architecture uses deep RL to infer actions from semantic and depth information, flying the robot towards secure areas, while respecting safety constraints. Thanks to our robust training strategy and the choice of these mid-level representations as input to the RL agent, we show that our policy can directly transfer to the real world, without the need for any additional fine-tuning. In a series of challenging experiments both in simulation and with a real platform, we demonstrate that our planner guides a rotorcraft UAV to a safe landing spot up to 1.5 times faster and with double success rate than the state of the art (including a commercially available solution), paving the way towards realistically deployable UAVs.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Autonomous Emergency Landing for Multicopters using Deep Reinforcement Learning

Bartolomei

Kompis

Teixeira

et al. 2022

2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

View full text Add to dashboard Cite

show abstract

“…The last decade has brought numerous breakthroughs in the development of autonomous robots which is evident from the manufacturing and service industries. More interesting are the advances that are essential enablers of several innovative applications such as robot-assisted surgery (Tewari et al, 2002), transportation (Thrun, 1995), environmental monitoring (Valada et al, 2012), planetary exploration (Toupet et al, 2020) and disaster relief (Mittal et al, 2019). Novel machine learning algorithms accompanied by the boost in computational capacity and availability of large annotated datasets have primarily fostered the progress in this field.…”

Section: Introductionmentioning

confidence: 99%

From Learning to Relearning: A Framework for Diminishing Bias in Social Robot Navigation

Hurtado,

Londoño,

Valada

2021

Preprint

Self Cite

View full text Add to dashboard Cite

The exponentially increasing advances in robotics and machine learning are facilitating the transition of robots from being confined to controlled industrial spaces to performing novel everyday tasks in domestic and urban environments. In order to make the presence of robots safe as well as comfortable for humans, and to facilitate their acceptance in public environments, they are often equipped with social abilities for navigation and interaction. Socially compliant robot navigation is increasingly being learned from human observations or demonstrations. We argue that these techniques that typically aim to mimic human behavior do not guarantee fair behavior. As a consequence, social navigation models can replicate, promote, and amplify societal unfairness such as discrimination and segregation. In this work, we investigate a framework for diminishing bias in social robot navigation models so that robots are equipped with the capability to plan as well as adapt their paths based on both physical and social demands. Our proposed framework consists of two components: learning which incorporates social context into the learning process to account for safety and comfort, and relearning to detect and correct potentially harmful outcomes before the onset. We provide both technological and sociological analysis using three diverse case studies in different social scenarios of interaction. Moreover, we present ethical implications of deploying robots in social environments and propose potential solutions. Through this study, we highlight the importance and advocate for fairness in human-robot interactions in order to promote more equitable social

show abstract

Continual SLAM: Beyond Lifelong Simultaneous Localization and Mapping Through Continual Learning

Vödisch

Cattaneo

Burgard

et al. 2023

Springer Proceedings in Advanced Robotics

Self Cite

View full text Add to dashboard Cite

Visual odometry is a fundamental task for many applications on mobile devices and robotic platforms. Since such applications are oftentimes not limited to predefined target domains and learning-based vision systems are known to generalize poorly to unseen environments, methods for continual adaptation during inference time are of significant interest. In this work, we introduce CoVIO for online continual learning of visual-inertial odometry. CoVIO effectively adapts to new domains while mitigating catastrophic forgetting by exploiting experience replay. In particular, we propose a novel sampling strategy to maximize image diversity in a fixed-size replay buffer that targets the limited storage capacity of embedded devices. We further provide an asynchronous version that decouples the odometry estimation from the network weight update step enabling continuous inference in real time. We extensively evaluate CoVIO on various real-world datasets demonstrating that it successfully adapts to new domains while outperforming previous methods. The code of our work is publicly available at http://continual-slam.cs.uni-freiburg.de.

show abstract

Vision-Based Autonomous UAV Navigation and Landing for Urban Search and Rescue

Cited by 22 publications

References 21 publications

Autonomous Emergency Landing for Multicopters using Deep Reinforcement Learning

Autonomous Emergency Landing for Multicopters using Deep Reinforcement Learning

From Learning to Relearning: A Framework for Diminishing Bias in Social Robot Navigation

Continual SLAM: Beyond Lifelong Simultaneous Localization and Mapping Through Continual Learning

Contact Info

Product

Resources

About