Analysis of the new market trends of UAVS for safe BVLOS operations

UAVs are currently conquering the skies as prominent tools for various data-intensive applications, in the economic, transport, military and civil sector. While initially found application in the military sector, technology progression allowed them to enter the recreational sector and are now gaining ground in the fringes of the commercial environment. In parallel, technical components and subsystems that are application-optimised focus on highly automated drones, benefit from expertise in other domains, especially when it comes to Electronic Components and Systems (ECS), such as the automotive one, to operate beyond the visual line of sight (BVLOS) with a rather high degree of autonomy. Such technological developments, as well as currents trends and societal needs have opened the way for an unparalleled expansion in the use of UAS for a great number of applications, where humans cannot reach or are unable to perform in a timely and efficient manner. This work aims to present a in-depth analysis of the current market trends that shape the existing landscape for the development of safe and reliable BVLOS operations.


Introduction
The drone industry in a worldwide scale has an exponential growth throughout the years, especially due to the constant advancements in avionics, navigation, propulsion, manufacturing, artificial intelligence, telecommunications, and computing capabilities of devices.Drones are Unmanned Aircraft Systems (UAS), also known as Unmanned Aerial Vehicles (UAVs), that is, aircraft without pilot or passengers, typically remote-controlled, maybe with a degree of autonomy [2].While initially found in the military sector, technology progression allowed them to enter the recreational sector and are now gaining ground in the fringes of the commercial environment [18].In parallel, technical components and subsystems that are applicationoptimised focus on highly automated UAS, they benefit from expertise in other domains, especially when it comes to Electronic Components and Systems (ECS), such as the automotive one.This market demands practices, standards and regulations and the work to deliver those is already well underway in SESAR3 [5].Innovations so far are reflected on drones equipped with functionality for avoiding any obstacles and flying from point A to point B. However, the continuous advances in the field open the gates for numerous versatile applications.Industries will anticipate significantly straightforward benefits from taking drones several steps forward, ensuring the digital sovereignty of the relevant European technology.Indicatively, autonomous drones at work can relieve humans from working in hazardous environments, reducing injuries caused by industrial accidents, perform frequent and efficient inspections, provide dynamic security, ensure access to high-frequency aerial data, utilize edge computing in an efficient manner, enable rapid response during emergencies, as well as enable object building maintenance and consumer market for house and garden keeping.Economically they get more interesting when they are able to enter higher volume market applications [25].
The deployment of UAS is witnessing an unparalleled expansion in recent years in various application domains.These sophisticated tools are currently conquering the skies as prominent tools for various data intensive applications in diverse sectors.Technology advancements in sensors and communication technologies have opened the way for new applications that will further boost their expansion, leading in turn to a growing interest among business leaders, policy makers and the public, which and has created more business opportunities.BVLOS operations in particular have gained significant attention in recent years, for their higher degrees of intelligence that brings several benefits such as greater efficiency and minimization of human risks.One of the greatest challenges for the greater adoption of UAV operation is safety and the resulting trust, which in turn represents the prominent driving force to reach a high technology acceptance from users.Performance, energy efficiency, security are typical examples of requirements that need to be balanced when the drones are meant to execute computationally intensive operations for their navigation and execution of alternate tasks.Such tasks need to be take place in a more dynamic execution environment that will provide the flexibility to allocate and deallocate resources at runtime depending on the task objectives, maintaining the necessary trade-off between the conflicting requirements.What's more, BVLOS sets a number of complex requirements to the UAVs operation, such as performance and energy efficiency [26].With this being said, the emergence of recent market trends of BVLOS operations and their influence to the integration of safe and autonomous UAVs operations will be the main driver in the improvement of multiple applications.
This work aims at presents an overview of the current market size and growth of the UAV industry by identifying emerging trends related to BVLOS capabilities, while it highlights the current technological advancements and innovations that are the main drivers for potential growth in BVLOS operations.The remainder of the paper is organized as follows.In section 2 that follows we discuss the current status of the BVLOS market, as well as its future potential.Section 3 presents an overview of the technologies that drive the UAVs market.In section 4 we discuss the emerging future scenarios that have opened the way for new applications in the area of BVLOS operations.Finally in section 5 we present the conclusions of this work, as well as the future directions.

Background
As UAVs are a rapidly evolving technology with significant market potential, their use can bring significant advantages, such as high mobility, easy deployment, capability to reach places that are inaccessible to humans.Particularily BVLOS capabilities are becoming recently a pivotal aspect for the drone industry.Compared to VLOS where the UAV should always be within the inobstructed view of the pilot, or EVLOS flights, where the operator may partly rely on critical flight information from remote observers that are in visual line with the UAV, BVLOS bring several benefits related to autonomy, but raises new challenges.Extended autonomy in BVLOS capabilities enables a drone to cover far greater distances, with a lower cost and a reduced risk to human life.On the other side, it creates a new landscape for communication, navigation, and flight under varying conditions [6].Accomplishing safe and efficient BVLOS operations can be a challenging and complex process that depends on several factors for instance, transmission and reception of data, execution of software functions, and environmental conditions, such as higher wind speed.Moreover, UAVs may consume large amounts of propulsion power to accelerate, maintain airborne or perform hovering tasks at different altitudes [18].In recent years, a number of works have focused on improving the performance and robustness of UAV deployment by exploring different optimization criteria, while he high demand of access to airspace in urban areas will require reinforced solutions and abilities for UAS operations, including navigation accuracy, management of failure and detect and avoid (DAA) capabilities [23].
The technology dimension can, however, only enable the compliance of operations without the regulatory framework which must also be defined.The high demand of access to airspace in urban areas will require the establishment of specific requirements and procedures to ensure safety in the air but also protect people on the ground and limit the impact of drone traffic on environment (emissions, noise, visual pollution).These operations will rely on secure and robust communication systems which can cope with large number of simultaneous operations as well as interference issues caused by large buildings.Drones operating in urban areas will have to comply with reinforced requirements and abilities, including navigation accuracy, management of failure and detect and avoid (DAA) capabilities.In this context, U-space will play a major role in managing the drone traffic and ensuring the interface with all the actors involved in the broader drone ecosystem.With a set of services and procedures for supporting complex UAS operations, U-space supports the integration of safe and efficient UAS missions in all types of operational environments, including urban areas enable complex drone operations with a high degree of automation [31].

Technologies driving the market growth of BVLOS
Over the past few years, drone industry investments have been progressively reaching new records.However, the year 2022 finally saw an end to the exponential growth, and the total value of investments decreased by 39% compared to the previous year.The market growth is attributed to the increasing enterprise application of drones across various industry verticals.Several drone manufacturers are continually testing, inventing, and upgrading solutions for diverse markets.Besides, the integration of modern technologies in commercial drones to deliver enhanced solutions is opening new growth opportunities for the commercial drone industry.The business use cases of commercial drones have expanded significantly over the past few years [10].
The global total drone market can be divided into two main categories: the commercial and the recreational drone market.The use of drones for commercial purposes means their deployment in commercial sectors such as energy, construction, agriculture, or transport & warehousing.Drones for commercial purposes are thus used either to generate direct income or to conduct business more efficiently.On the other hand, the recreational drone market includes consumers such as children, adults, and semi-professionals who use drones for recreational or hobby purposes such as aerial filming and photography.Those who use drones to earn money for a (small) additional income also belong to the category of the recreational drone market in Figure 1.
Enabling successful and safe operations with an autonomous, intelligent and safe nature is the core of civilian drone market potential and this will require the availability of a variety of technologies developed.As UAVs are currently conquering the skies for various data intensive applications, in the economic, transport, military and civil sector, technology advancements in sensors and communication technologies have opened the way for new applications that will further boost their use.In the following section we present an updated list of the state of the art technologies that have been identified as drivers for future drone growth.

Sensors, actuators and perception for airborne automated systems
A new generation of dedicated sensors has been conceived over the last few years with main focus on environmental detection and perception of autonomous, intelligent and safe drones operations.Current drone detection and movement characterization solutions are based on visual, acoustic, radio frequency (RF), optoelectronic , and multimodal sensors.[8].Vision aided inertial navigation techniques independent of active sensing like lidar, radar, or ultrasonic, and independent of GNSS coverage experienced a significant advancement in the autonomous, intelligent and safes community.
Environmental perception in autonomous systems typically utilizes combinations of several algorithmic building blocks, such as point cloud segmentation, object detection, multi-object tracking, object classification, semantic segmentation and occupancy grid filtering [4].Different sensor modalities and configurations have their respective strengths and weaknesses.For instance, radar sensors can measure velocities directly but provide a very sparse measurement of the environment.Lidar sensors provide dense point clouds but cannot directly measure velocities.Mono camera sensors provide very rich visual cues for classification but lack distance information.The goal of sensor fusion is to combine data from different sources to increase the overall quality and quantity of the data.Therefore, sensor fusion is a natural approach to environmental perception, as it can combine the strengths of different sensor modalities while compensating for their drawbacks at the same time.Roughly speaking, there are three popular types of approaches for sensor fusion: statistical approaches, probabilistic approaches, and artificial intelligence (AI) based approaches [12].

Path planning for UAVs
Path planning for UAVs typically involves the determination of path from an initial point to a destination point, without the possibility of a collision with any existing obstacles [1]. .It is an umbrella term that involves Motion planning, which refers to path optimization in terms of length and minimum turning angle, Trajectory planning, which denotes the velocity and kinematics of the UAV and Navigation which refers to the controlling and monitoring of the UAVs movement from one place to another.Path planning methods can be classified in two categories, those that are based on a sampling of the search space and those that employ artificial intelligence (AI) to find a solution with respect to the representation of the environment [1].Various recent works have explored different novel path planning strategies such as evolutionary or Reinforcement Learning (RL) algorithms for UAV swarms [28,30,32].Advanced aerial autonomous, intelligent and safe navigation is gaining popularity in more complex applications requiring specific navigation algorithms with multimodal path planning optimization.Taking into account only simple optimization criteria (energy consumption, path length, object avoidance) for navigation algorithms is insufficient in advanced autonomous, intelligent and safe drone applications.Such applications are complex object inspections (total coverage optimization), interaction with other objects in the environment (optimization for a swarm or dynamical objects), or even delivery applications (risk minimization, path optimization for connectivity coverage), etc. Due to this fact, the complexity of configuration space consists of higher-order variables taking into account application, vehicle dynamic, and even environment (wind, etc.) constraints, which leads to the higher complexity of optimization criteria [13].
Graph-based shortest path finding algorithms, such as the Dijkstra's algorithm [9] and its variations [7], the A* path search algorithm [15], or the Fast Marching (FM) [29] Potential Fields [20].Probabilistically complete or sampling-based methods based on exploring Random Tree family [21] methods, have been extensively implemented for solving path planning problems.However, their complexity rises with the complexity of state space and mainly dimension of configuration space.In recent years, Reinforcement learning (RL) techniques and particularly Deep reinforcement learning (DRL) [3] are becoming popular in UAV navigation.Such methods employ on training a neural network using an RL strategy to learn from the results of past actions and uses environmental feedback as the input for path planning [16].

Federated Machine Learning with advanced Communication
Federated Machine Learning (FedML) can be used in a variety of ways in the context of unmanned aerial vehicles such as for better route planning, or target recognition.In contrast to other predecessor technology of machine learning (ML), FedML offers a solution for high data security and privacy, as offers training of the a model without sharing the corresponding data, while calculations are performed on the end device itself.
However, the use of FedML for drones and the like in practice is not yet widespread, despite its high potential.A reason for this might be that the communication of parameters is based on wireless networks which rely on ground-based infrastructure and are prone to transmission delays [34].This is particularly problematic in industrial drone operating scenarios characterized by non-ideal conditions such as harsh environments (e.g., open sea).For drone communication in general, no standard for the actual communication link has evolved yet.Possible communication technologies like the fourth and fifth generation of cellular network technology (4G/ 5G) have strong disadvantages since they are not fully rolled out yet (5G) and generally rely on ground-based infrastructure.In infrastructurally poorly developed environments, other communication standards that are independent of ground-based infrastructure are more suitable.Drone to anything (D2X) communication, for example, already shows its potential as a robust communication link based on the IEEE 802.11p communication technology.IEEE 802.11p was designed for vehicle-to-vehicle and vehicle-to-infrastructure communication in the automotive area.It features a low latency ad-hoc-network communication, which is designed for high relative speeds between transmitters and receivers and even multiple reflections of the radio wave by surrounding buildings or other obstacles are no issue.UAV surveilance and coordinate exchange is also critical for BVLOS operations, particularly for the Future Unmanned Traffic Management (UTM) systems.As a solution, Automatic Dependent Surveillance-Broadcast (ADS-B) technology is widely adopted for Beyond Visual Line of Sight (BVLOS) drone operations.ADS-B technology is a surveillance technology in which the aircraft determines its position based on global positioning system (GPS) data rather than radar, allowing the ATC to monitor the location every 500 ms [14,22].In recent years, this technology has paved the way for transforming the drone industry, enabling faster and more efficient operations for a variety of applications.However, the safe and effective implementation of ADS-B technologies for BVLOS drone operations brings several challenges and requires a regulatory framework that ensures the safe integration of drones into the national airspace.What's more integrating ADS-B technology into UAVs can be challenging due to size, weight, and power constraints, and thus more lightweight and compact ADS-B transponders should be developed [24].

Future UAV Scenarios
Technology advancements in sensors and communication technologies have opened the way for new applications that will further boost the use of UAVs and address the demands of more challenging operational scenarios.The future of UAV technologies holds great promise, with ongoing advancements expected in various aspects of design, capabilities, and applications.This section presents the key areas where future UAV developments are anticipated.

Collaborative drones
In recent years the emerging scenario of collaborative UAVs or swarm UAVs has been introduced for tackling the complexity of BVLOS missions.Inspired by the natural formations of creatures, such as the flock of birds or the schools of fish, a UAV swarm consists of a number of aerial vehicles with restricted autonomy and communication ability that operate together in a synergistic mode in order to perform cooperative exploration and mapping tasks, such as data collection, path planning, or co-mapping generation and validation [25,11].UAV swarms are considered unique instruments for monitoring and exploration tasks due to their autonomy and efficiency [27].Furthermore while multi-UAV collaborative operations offer higher scalability, more limited mission execution times, reduction in risk due to redundancy, possibility of multi-modal data capturing and overall improvement in the performance for example in time-critical operations [19].

Data fusion at the edge
The need to process data close to the source in real-time is becoming more common in many UAS applications nowadays [33].This need brings the topic of edge computing directly to the heart of flight controllers of current UAS systems.Different edge computing capable devices are used depending on the application complexity, considering processing power, energy consumption, and type of processed data.Distributed and multi-modal data fusion at the edge level can significantly improve the data quality in a distributed environment.One of the main advantages of edge computing devices on UAS is to reduce data flow during missions and gather only essential data from the environment in real-time, reducing needed bandwidth and storage capacity onboard.In many applications, complete data processing is done after the flight from gathered data during a mission.Though UAS applications use such edge computing devices, these are often general-purpose devices not specifically designed for UAS [17].

Conclusion
Recent technological developments, trends and societal needs have opened the way for an unparallel expansion in the use of UAS.This work has investigated the emerging key areas, that currently facilitate the market size and growth of the UAV industry in urbanised environments.These are shaped by a convergence of technological breakthroughs, regulatory developments, and the expanding industry adoption.Despite the promising future, the widespread integration of UAVs will require continuous advancements in technology, regulatory frameworks, and public acceptance to ensure safety, privacy, and ethical use.As these challenges are addressed, the full potential of UAVs will be realized, contributing to a more connected, efficient, and sustainable future.