Search

This work focuses on the efficiency improvement of grid-based Coverage Path Planning (CPP) methodologies in real-world applications with UAVs. While several sophisticated approaches are met in literature, grid-based methods are not commonly used in real-life operations. This happens mostly due to the error that is introduced during the region’s representation on the grid, a step mandatory for such methods, that can have a great negative impact on their overall coverage efficiency. A previous work on UAVs’ coverage operations for remote sensing, has introduced a novel optimization procedure for finding the optimal relative placement between the region of interest and the grid, improving the coverage and resource utilization efficiency of the generated trajectories, but still, incorporating flaws that can affect certain aspects of the method’s effectiveness. This work goes one step forward and introduces a CPP method, that provides three different ad-hoc coverage modes: the Geo-fenced Coverage Mode, the Better Coverage Mode and the Complete Coverage Mode, each incorporating features suitable for specific types of vehicles and real-world applications. For the design of the coverage trajectories, user-defined percentages of overlap (sidelap and frontlap) are taken into consideration, so that the collected data will be appropriate for applications like orthomosaicing and 3D mapping. The newly introduced modes are evaluated through simulations, using 20 publicly available benchmark regions as testbed, demonstrating their stenghts and weaknesses in terms of coverage and efficiency. The proposed method with its ad-hoc modes can handle even the most complex-shaped, concave regions with obstacles, ensuring complete coverage, no-sharp-turns, non-overlapping trajectories and strict geo-fencing. The achieved results demonstrate that the common issues encountered in grid-based methods can be overcome by considering the appropriate parameters, so that such methods can provide robust solutions in the CPP domain.

This paper is an initial endeavor to bridge the gap between powerful Deep Reinforcement Learning methodologies and the problem of exploration/coverage of unknown terrains. Within this scope, MarsExplorer, an openai-gym compatible environment tailored to exploration/coverage of unknown areas, is presented. MarsExplorer translates the original robotics problem into a Reinforcement Learning setup that various off-the-shelf algorithms can tackle. Any learned policy can be straightforwardly applied to a robotic platform without an elaborate simulation model of the robot's dynamics to apply a different learning/adaptation phase. One of its core features is the controllable multi-dimensional procedural generation of terrains, which is the key for producing policies with strong generalization capabilities. Four different state-of-the-art RL algorithms (A3C, PPO, Rainbow, and SAC) are trained on the MarsExplorer environment, and a proper evaluation of their results compared to the average human-level performance is reported. In the follow-up experimental analysis, the effect of the multi-dimensional difficulty setting on the learning capabilities of the best-performing algorithm (PPO) is analyzed. A milestone result is the generation of an exploration policy that follows the Hilbert curve without providing this information to the environment or rewarding directly or indirectly Hilbert-curve-like trajectories. The experimental analysis is concluded by evaluating PPO learned policy algorithm side-by-side with frontier-based exploration strategies. A study on the performance curves revealed that PPO-based policy was capable of performing adaptive-to-the-unknown-terrain sweeping without leaving expensive-to-revisit areas uncovered, underlying the capability of RL-based methodologies to tackle exploration tasks efficiently. The source code can be found at: https://github.com/dimikout3/MarsExplorer.

Part 14: Multi Agent Systems; International audience; The current study focuses on the problem of continuously tracking a dynamically evolving $$CH_4$$CH4 plume utilizing a mutually built consensus by heterogeneous sensing platforms: mobile and static sensors. Identifying the major complexities and emergent dynamics (leakage source, intensity, time) of such problem, a distributed, multi-agent, optimization algorithm was developed and evaluated in an indoor continuous plume-tracking application (where reaction time is critical due to the limited volume available for air saturation by the $$CH_4$$CH4 dispersion). The high-fidelity ANSYS Fluent suite realistic simulation environment was used to acquire the gas diffusion evolution through time. The analysis of the simulation results indicated that the proposed algorithm was capable of continuously readapting the mobile sensing platforms formation according to the density and the dispersed volume plume; combining additive information from the static sensors. Moreover, a scalability analysis with respect to the number of mobile platforms revealed the flexibility of the proposed algorithm to different numbers of available assets.

Political instability around the world continues to place a significant emphasis on border control. Monitoring these borders requires persistent surveillance in a variety of remote, hazardous and hostile environments. While recent developments in autonomous and unmanned systems promise to provide a new generation of tools to assist in border control missions, the complexity of designing, testing and operating large-scale systems limits their adoption.

Cross-border crime utilizes recent advanced systems to perform their illegal activities. Innovative sensory systems and specialized equipment are examples that were used for trafficking of human and of various illicit materials. The increasing challenges that border personnel must resolve require the usage of recent technological advances as well. Thus, the utilization of pioneer technologies seems imperative to precede technologically organized crime. Towards this objective, the introduction of unmanned vehicles (UxV) and the advances of relevant sub-systems have created a new solution to fight cross-border crime. Utilizing a combination of UxVs enriched with enhanced detection capabilities comprises an effective solution. The chapter will introduce and present the capability of an autonomous navigation system by exploiting swarm intelligence principles towards simplifying the overall operation. Computer vision advances and semantic enrichment of the acquired information are incorporated to deliver cutting-edge technologies. The described architecture and services can provide a complete solution for optimal border surveillance and increased situation awareness.

Over the last years, an intensified interest has been shown in many studies for precision agriculture. Unmanned Aircraft Systems (UASs) are capable of solving a plethora of surveying tasks due to their flexibility, independence and customization. The incorporation of UASs remote sensing in precision agriculture enhances the abilities of crop mapping, management and identification through vegetation indices. In addition to this, different image analysis and computer vision processes were adopted trying to facilitate field operations in cooperation with human intervention to enhance the overall performance. In this paper, we present a practically oriented application on vineyards towards an integrated low-cost system which utilizes Spiral-STC (Spanning Tree Coverage) algorithm as a Coverage Path Planning (CPP) method. Based on the resulted flight campaign, UAV images were collected, and the incorporated image analysis processes finally extract vegetation knowledge. Also, geo referenced orthophotos and computer vision applications complete the generated oversight of the field. These supportive tools provide farmers with useful information (crop health indicators, weather predictions) letting them extrapolate knowledge and identify crop irregularities.

This article investigates the problem of deploying a swarm of UAVs equipped with gas sensors for industrial remote gas-plume sensing. This setup’s objective is to continuously adjust the swarm formation to maximize the combined perception for the dynamically evolved plume’s cloud, focusing around areas with the highest concentration/intensity. Initially, such a setup is formulated into an optimization problem, the solution of which could be acquired by the maximization of an appropriately defined objective function. Due to the model-free approach, this objective function’s analytical form is not available, prohibiting standard gradient descent methodologies. To this end, a tracking algorithm is developed and studied, which operates in a distributed manner and enables the UAV swarm to build a common consensus dynamically, during the evolution of the leakage phenomenon. The overall performance is tested in a simulative yet realistic environment using ANSYS Fluent suite, considering a simultaneous gas-leak incident at two different points. Aside from the standalone evaluation study, the proposed gas-plume tracking scheme is able to outperform a state-of-the-art alternative algorithm, namely Efficient Global Optimization (EGO), in various simulation setups, deploying a different number of UAVs on the field.

The introduction of Unmanned vehicles (UxVs) in the recent years has created a new security field that can use them as both a potential threat as well as new technological weapons against those threats. Dealing with these issues from the counter-threat perspective, the proposed architecture project focuses on designing and developing a complete system which utilizes the capabilities of multiple UxVs for surveillance objectives in different operational environments. Utilizing a combination of diverse UxVs equipped with various sensors, the developed architecture involves the detection and the characterization of threats based on both visual and thermal data. The identification of objects is enriched with additional information extracted from other sensors such as radars and RF sensors to secure the efficiency of the overall system. The current prototype displays diverse interoperability concerning the multiple visual sources that feed the system with the required optical data. Novel detection models identify the necessary threats while this information is enriched with higher-level semantic representations. Finally, the operator is informed properly according to the visual identification modules and the outcomes of the UxVs operations. The system can provide optimal surveillance capacities to the relevant authorities towards an increased situational awareness.

This research develops a new on-line trajectory planning algorithm for a team of Autonomous Underwater Vehicles (AUVs). The goal of the AUVs is to cooperatively explore and map the ocean seafloor. As the morphology of the seabed is unknown and complex, standard non-convex algorithms perform insufficiently. To tackle this, a new simulation-based approach is proposed and numerically evaluated. This approach adapts the Parametrized Cognitive-based Adaptive Optimization (PCAO) algorithm. The algorithm transforms the exploration problem to a parametrized decision-making mechanism whose real-time implementation is feasible. Upon that transformation, this scheme calculates off-line a set of decision making mechanism’s parameters that approximate the - non-practically feasible - optimal solution. The advantages of the algorithm are significant computational simplicity, scalability, and the fact that it can straightforwardly embed any type of physical constraints and system limitations. In order to train the PCAO controller, two morphologically different seafloors are used. During this training, the algorithm outperforms an unrealistic optimal-one-step-ahead search algorithm. To demonstrate the universality of the controller, the most effective controller is used to map three new morphologically different seafloors. During the latter mapping experiment, the PCAO algorithm outperforms several gradient-descent-like approaches.

This paper deals with the problem of autonomous exploration of unknown areas using teams of Autonomous X Vehicles (AXVs)—with X standing for Aerial, Underwater or Sea-surface—where the AXVs have to autonomously navigate themselves so as to construct an accurate map of the unknown area. Such a problem can be transformed into a dynamic optimization problem which, however, is NP-complete and thus infeasible to be solved. A usual attempt is to relax this problem by employing greedy (optimal one-step-ahead) solutions which may end-up quite problematic. In this paper, we first show that optimal one-step-ahead exploration schemes that are based on a transformed optimization criterion can lead to highly efficient solutions to the multi-AXV exploration. Such a transformed optimization criterion is constructed using both theoretical analysis and experimental investigations and attempts to minimize the “disturbing” effect of deadlocks and nonlinearities to the overall exploration scheme. As, however, optimal one-step-ahead solutions to the transformed optimization criterion cannot be practically obtained using conventional optimization schemes, the second step in our approach is to combine the use of the transformed optimization criterion with the cognitive adaptive optimization (CAO): CAO is a practicably feasible computational methodology which adaptively provides an accurate approximation of the optimal one-step-ahead solutions. The combination of the transformed optimization criterion with CAO results in a multi-AXV exploration scheme which is both practically implementable and provides with quite efficient solutions as it is shown both by theoretical analysis and, most importantly, by extensive simulation experiments and real-life underwater sea-floor mapping experiments in the Leixes port, Portugal.

This paper deals with the path planning problem of a team of mobile robots, in order to cover an area of interest, with prior-defined obstacles. For the single robot case, also known as single robot coverage path planning (CPP), an 𝓞(n) optimal methodology has already been proposed and evaluated in the literature, wherenis the grid size. The majority of existing algorithms for the multi-robot case (mCPP), utilize the aforementioned algorithm. Due to the complexity, however, of the mCPP, the best the existing mCPP algorithms can perform is at most 16 times the optimal solution, in terms of time needed for the robot team to accomplish the coverage task, while the time required for calculating the solution is polynomial. In the present paper, we propose a new algorithm which converges to the optimal solution, at least in cases where one exists. The proposed technique transforms the original integer programming problem (mCPP) into several single-robot problems (CPP), the solutions of which constitute the optimal mCPP solution, alleviating the original mCPP explosive combinatorial complexity. Although it is not possible to analytically derive bounds regarding the complexity of the proposed algorithm, extensive numerical analysis indicates that the complexity is bounded by polynomial curves for practically sized inputs. In the heart of the proposed approach lies the DARP algorithm, which divides the terrain into a number of equal areas each corresponding to a specific robot, so as to guaranteecomplete coverage,non-backtracking solution,minimum coverage path, while at the same timedoes not need any preparatory stage(video demonstration and standalone application are available on-linehttp://tinyurl.com/DARP-app).

In this paper, we explore, extend and simplify the localization of the description ability of the well-established MPEG-7 (Scalable Colour Descriptor (SCD), Colour Layout Descriptor (CLD) and Edge Histogram Descriptor (EHD)) and MPEG-7-like (Color and Edge Directivity Descriptor (CEDD)) global descriptors, which we call the SIMPLE family of descriptors. Sixteen novel descriptors are introduced that utilize four different sampling strategies for the extraction of image patches to be used as points of interest. Designing with focused attention for content-based image retrieval tasks, we investigate, analyse and propose the preferred process for the definition of the parameters involved (point detection, description, codebook sizes and descriptors’ weighting strategies). The experimental results conducted on four different image collections reveal an astonishing boost in the retrieval performance of the proposed descriptors compared to their performance in their original global form. Furthermore, they manage to outperform common SIFT- and SURF-based approaches while they perform comparably, if not better, against recent state-of-the-art methods that base their success on much more complex data manipulation.

Accurate maps are essential in the case of robot teams, so that they can operate autonomously and accomplish their tasks efficiently. In this work we present an approach which allows the generation of detailed maps, suitable for robot navigation, from a mesh of sparse points using Cellular Automata and simple evolutions rules. The entire map area can be considered as a 2D Cellular Automaton (CA) where the value at each CA cell represents the height of the ground in the corresponding coordinates. The set of measurements form the original state of the CA. The CA rules are responsible for generating the intermediate heights among the real measurements. The proposed method can automatically adjust its rules, so as to encapture local morphological attributes, using a pre-processing procedure in the set of measurements. The main advantage of the proposed approach is the ability to maintain an accurately reconstruction even in cases where the number of measurements are significant reduced. Experiments have been conducted employing data collected from two totally different real-word environments. In the first case the proposed approach is applied, so as to build a detailed map of a large unknown underwater area in Oporto, Portugal. The second case concerns data collected by a team of aerial robots in real experiments in an area near Zurich, Switzerland and is also used for the evaluation of the approach. The data collected, in the two aforementioned cases, are extracted using different kind of sensors and robots, thus demonstrating the applicability of our approach in different kind of devices. The proposed method outperforms the performance of other well-known methods in literature thus enabling its application for real robot navigation.

In this paper we propose and evaluate a new technique that localizes the description ability of the well established MPEG-7 and MPEG-7-like global descriptors. We employ the SURF detector to define salient image patches of blob-like textures and use the MPEG-7 Scalable Color (SC), Color Layout (CL) and Edge Histogram (EH) descriptors and the global MPEG-7-like Color and Edge Directivity Descriptor (CEDD), to produce the final local features' vectors. In order to test the new descriptors in the most straightforward fashion, we use the Bag-Of-Visual-Words framework for indexing and retrieval. The experimental results conducted on two different benchmark databases with varying codebook sizes, revealed an astonishing boost in the retrieval performance of the proposed descriptors compared both to their own performance (in their original form) and to other state-of-the-art methods of local and global descriptors. Open-source implementation of the proposed descriptors is available in c#, Java and MATLAB.

This paper proposes a novel mission planning platform, capable of efficiently deploying a team of UAVs to cover complex-shaped areas, in various remote sensing applications. Under the hood lies a novel optimization scheme for grid-based methods, utilizing Simulated Annealing algorithm, that significantly increases the achieved percentage of coverage and improves the qualitative features of the generated paths. Extensive simulated evaluation in comparison with a state-of-the-art alternative methodology, for coverage path planning (CPP) operations, establishes the performance gains in terms of achieved coverage and overall duration of the generated missions. On top of that, DARP algorithm is employed to allocate sub-tasks to each member of the swarm, taking into account each UAV's sensing and operational capabilities, their initial positions and any no-fly-zones possibly defined inside the operational area. This feature is of paramount importance in real-life applications, as it has the potential to achieve tremendous performance improvements in terms of time demanded to complete a mission, while at the same time it unlocks a wide new range of applications, that was previously not feasible due to the limited battery life of UAVs. In order to investigate the actual efficiency gains that are introduced by the multi-UAV utilization, a simulated study is performed as well. All of these capabilities are packed inside an end-to-end platform that eases the utilization of UAVs' swarms in remote sensing applications. Its versatility is demonstrated via two different real-life applications: (i) a photogrametry for precision agriculture and (ii) an indicative search and rescue for first responders missions, that were performed utilizing a swarm of commercial UAVs. The source code can be found at: https://github.com/savvas-ap/mCPP-optimized-DARP, An implementation of the mCPP methodology introduced in this work, as well as a link for a demonstrative video and a link for a fully functional, on-line hosted instance of the presented platform can be found here: https://github.com/savvas-ap/mCPP-optimized-DARP

In order to sufficiently protect active personnel and physical environment from hazardous leaks, recent industrial practices integrate innovative multi-modalities so as to maximize response efficiency. Since the early detection of such incidents portrays the most critical factor for providing efficient response measures, the continuous and reliable surveying of industrial spaces is of primary importance. Current study develops a surveying mechanism, utilizing a swarm of heterogeneous aerial mobile sensory platforms, for the continuous monitoring and detection of CH4 dispersed gas plumes. In order to timely represent the CH4 diffusion progression incident, the research concerns a simulated indoor, geometrically complex environment, where early detection and timely response are critical. The primary aim was to evaluate the efficiency of a novel multi-agent, closed-loop, algorithm responsible for the UAV path-planning of the swarm, in comparison with an efficient a state-of-the-art path-planning EGO methodology, acting as a benchmark. Abbreviated as Block Coordinate Descent Cognitive Adaptive Optimization (BCD-CAO) the novel algorithm outperformed the Efficient Global Optimization (EGO) algorithm, in seven simulation scenarios, demonstrating improved dynamic adaptation of the aerial UAV swarm towards its heterogeneous operational capabilities. The evaluation results presented herein, exhibit the efficiency of the proposed algorithm for continuously conforming the mobile sensing platforms’ formation towards maximizing the total measured density of the diffused volume plume.

Current multi-AUV systems are far from being capable of fully autonomously taking over real-life complex situation-awareness operations. As such operations require advanced reasoning and decision-making abilities, current designs have to heavily rely on human operators. The involvement of humans, however, is by no means a guarantee of performance; humans can easily be over-whelmed by the information overload, fatigue can act detrimentally to their performance, properly coordinating vehicles actions is hard, and continuous operation is all but impossible. Within the European funded project NOPTILUS we take the view that an effective fully-autonomous multi-AUV concept/system, is capable of overcoming these shortcomings, by replacing human-operated operations by a fully autonomous one. In this paper, we present a new approach that is able to efficiently and fully-autonomously navigate a team of AUVs when deployed in exploration of unknown static and dynamic environments towards providing accurate static/dynamic maps of the environment. Additionally to achieving to efficiently and fully-autonomously navigate the AUV team, the proposed approach possesses certain advantages such as its extremely computational simplicity and scalability, and the fact that it can very straightforwardly embed and type of physical or other constraints and limitations (e.g., obstacle avoidance, nonlinear sensor noise models, localization fading environments, etc).

Recently, special attention has been paid in developing methodologies and systems for embedding autonomy within smart devices (Things). Moreover, as Things typically operate in an interconnected IoT ecosystem, autonomous operation must be performed in a cooperative fashion so the different Things coordinate their autonomous actions towards meeting high-level objectives and policies. Embedding Things with cooperative autonomy typically requires a tedious and costly effort not only during the original ecosystem deployment but throughout its lifetime. The current study describes CAO (Cognitive Adaptive Optimization)—and its distributed counterpart L4G-CAO (Local for Global Cognitive Adaptive Optimization)—which can overcome this shortcoming. CAO and L4G-CAO—which have recently been introduced and tested in a variety of IoT applications—can embed Things with cooperative autonomy in a plug-n-play fashion, i.e., without requiring the aforementioned tedious and costly effort. Results of the application of the aforementioned approaches in three different application domains (smart homes and districts, intelligent traffic systems and coordinated swarms of robots) are also presented. The presented results demonstrate the potential, of both approaches, to exploit the IoT automation functionalities in order to significantly improve the overall IoT performance without tedious effort.

This paper deals with the problem of retrieving the optimal path between two points inside an unknown environment, utilizing a robot-scouter. The vast majority of the path planning frameworks for an unknown environment focuses on the problem of navigating a robot, as soon as possible, towards a pre-specified location. As a result, the final followed path between the start and end location is not necessarily the optimal one, as the objective of the robot at each timestamp is to minimize its current distance to the desirable location. However, there are several real-life applications, like the one formulated in this paper, where the robot-scouter has to find the minimum path between two positions in an unknown environment, which is going to be used in a future phase. In principle, the optimal path can be guaranteed by a searching agent that adopts an A∗-like decision mechanism. In this paper, we propose a specifically-tailored variation (CIA∗) of the A∗ algorithm to the problem in hand. CIA∗ inherits the A∗ optimality and efficiency guarantees, while at the same time exploits the learnt formation of the obstacles, to on-line revise the heuristic evaluation of the candidate states. As reported in the simulation results, CIA∗ achieves an enhancement in the range of 20-50%, over the typical A∗, on the cells that have to be visited to guarantee the optimal path construction. An open-source implementation of the proposed algorithm along with a Matlab GUI are available1.