Multilayer Mapping Kit for Autonomous UAV Navigation

International Journal of Robotics and Automation Technology

The construction and maintenance of a robocentric map is key to high-level mobile robotic tasks like path planning and smart navigation. But the challenge of dynamic environment and huge amount of dense sensor data makes it hard to be implemented in a real-world application for long-term use. In this paper we present a novel mapping approach by incorporating semantic cuboid object detection and multi-view geometry information. The proposed system can precisely describe the incremental 3D environment in real-time and maintain a long-term map by extracting out moving objects. The representation of the map is a collection of sub-volumes which can be utilized to perform pose graph optimization to address the challenge of building a consistent and scalable map. These sub-volumes are first aligned by localization module and refined by fusing the active volumes using co-visible graph. With the proposed framework we can obtain the object-level constraints and propose a consistent obstacle m...

Three dimensional data is in great demand for disaster and environmental monitoring. In order to represent 3D space in details, it is indispensable to acquire 3D shape and texture together efficiently. However, they still lack a reliable, quick, cheap and handy method of acquiring three dimensional data of objects at higher resolution and accuracy in outdoors and moving environments. In this research, utilization of UAV (Unmanned Aerial Vehicle) is proposed. UAV is often used in military purpose, but also UAV is currently used in civil purpose such as mapping and disaster monitoring. UAV can fly lower altitude to get precise information and it is good for dangerous situation, however UAV has some limitations in terms of payload. It is difficult to mount aerial remote sensing sensors, so UAV mapping is still at the level of monitoring, not at the level of surveying. In this research, a combination of CCD cameras and a small (cheap) laser scanner with inexpensive IMU and GPS is proposed for an UAV borne 3D mapping system. Direct geo-referencing is achieved automatically using all the sensors without any ground control points. A new method of direct geo-referencing by the combination of bundle block adjustment and Kalman filter is proposed. This is the way of rendering objects with rich shape and detailed texture automatically by using an UAV.

ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2015

In the last few years, low-cost UAV systems have been acknowledged as an affordable technology for geospatial data acquisition that can meet the needs of a variety of traditional and non-traditional mapping applications. In spite of its proven potential, UAV-based mapping is still lacking in terms of what is needed for it to become an acceptable mapping tool. In other words, a well-designed system architecture that considers payload restrictions as well as the specifications of the utilized direct geo-referencing component and the imaging systems in light of the required mapping accuracy and intended application is still required. Moreover, efficient data processing workflows, which are capable of delivering the mapping products with the specified quality while considering the synergistic characteristics of the sensors onboard, the wide range of potential users who might lack deep knowledge in mapping activities, and time constraints of emerging applications, are still needed to be ...

arXiv (Cornell University), 2021

In this work we propose a holistic framework for autonomous aerial inspection tasks, using semantically-aware, yet, computationally efficient planning and mapping algorithms. The system leverages state-of-the-art receding horizon exploration techniques for next-best-view (NBV) planning with geometric and semantic segmentation information provided by state-of-the-art deep convolutional neural networks (DCNNs), with the goal of enriching environment representations. The contributions of this article are threefold, first we propose an efficient sensor observation model, and a reward function that encodes the expected information gains from the observations taken from specific view points. Second, we extend the reward function to incorporate not only geometric but also semantic probabilistic information, provided by a DCNN for semantic segmentation that operates in real-time. The incorporation of semantic information in the environment representation allows biasing exploration towards specific objects, while ignoring task-irrelevant ones during planning. Finally, we employ our approaches in an autonomous drone shipyard inspection task. A set of simulations in realistic scenarios demonstrate the efficacy and efficiency of the proposed framework when compared with the state-of-the-art.

For situations, where mapping is neither possible from high altitudes nor from the ground, we are developing an autonomous micro aerial vehicle able to fly at low altitudes in close vicinity of obstacles. This vehicle is based on a MikroKopter TM octocopter platform (maximum total weight: 5kg), and contains a dual frequency GPS board, an IMU, a compass, two stereo camera pairs with fisheye lenses, a rotating 3D laser scanner, 8 ultrasound sensors, a real-time processing unit, and a compact PC for on-board ego-motion estimation and obstacle detection for autonomous navigation. A high-resolution camera is used for the actual mapping task, where the environment is reconstructed in three dimensions from images, using a highly accurate bundle adjustment. In this contribution, we describe the sensor system setup and present results from the evaluation of several aspects of the different subsystems as well as initial results from flight tests.

arXiv (Cornell University), 2019

This work presents a modular architecture for simultaneous mapping and target driven navigation in indoors environments. The semantic and appearance stored in 2.5D map is distilled from RGB images, semantic segmentation and outputs of object detectors by convolutional neural networks. Given this representation, the mapping module learns to localize the agent and register consecutive observations in the map. The navigation task is then formulated as a problem of learning a policy for reaching semantic targets using current observations and the up-to-date map. We demonstrate that the use of semantic information improves localization accuracy and the ability of storing spatial semantic map aids the target driven navigation policy. The two modules are evaluated separately and jointly on Active Vision Dataset [1] and Matterport3D environments [3], demonstrating improved performance on both localization and navigation tasks.

IEEE Robotics and Automation Letters, 2020

Globally consistent dense maps are a key requirement for long-term robot navigation in complex environments. While previous works have addressed the challenges of dense mapping and global consistency, most require more computational resources than may be available on-board small robots. We propose a framework that creates globally consistent volumetric maps on a CPU and is lightweight enough to run on computationally constrained platforms. Our approach represents the environment as a collection of overlapping Signed Distance Function (SDF) submaps, and maintains global consistency by computing an optimal alignment of the submap collection. By exploiting the underlying SDF representation, we generate correspondencefree constraints between submap pairs that are computationally efficient enough to optimize the global problem each time a new submap is added. We deploy the proposed system on a hexacopter Micro Aerial Vehicle (MAV) with an Intel i7-8650U CPU in two realistic scenarios: mapping a large-scale area using a 3D LiDAR, and mapping an industrial space using an RGB-D camera. In the large-scale outdoor experiments, the system optimizes a 120x80 m map in less than 4 s and produces absolute trajectory RMSEs of less than 1 m over 400 m trajectories. Our complete system, called voxgraph, is available as open source 1 .

2010

This paper describes a scalable dense 3D reconstruction and navigation system suitable for real-time operation.

2007

The desire to use Unmanned Air Vehicles (UAVs) in a variety of complex missions has motivated the need to increase the autonomous capabilities of these vehicles. This research presents autonomous vision-based mapping and trajectory planning strategies for a UAV navigating in an unknown urban environment. It is assumed that the vehicle's inertial position is unknown because GPS in unavailable due

International Journal of Robotic Research, 2004

This paper addresses the problem of autonomous navigation in very large unstructured environments. A new Hybrid Metric Map (HYMM) structure is presented that combines feature maps with other metric representation in a consistent manner. The global feature map is partitioned into a set of connected Local Triangular Regions (LTRs), which provided a reference for a detailed multi-dimensional description of the environment. The HYMM framework permits the combination of efficient feature-based SLAM algorithms for localisation with, for example, occupancy grid maps for tasks such as obstacle avoidance, path planning or data association. This fusion of feature and grid maps has several complementary properties; for example, grid maps can assist data association and can facilitate the extraction and incorporation of new landmarks as they become identified from multiple vantage points. This paper also presents a path planning technique that efficiently maintains the estimated cost of traversing each LTR. The consistency of the SLAM algorithm is investigated with the introduction of exploration techniques to guarantee a certain measure of performance for the estimation process. Experimental results in outdoor environments are presented to demonstrate the performance of the algorithms proposed.