Chinese automakers are starting to equip electric cars with camera drones. For now, this drone integration is aimed at content creators who want to collect videos of themselves driving. These systems typically enable one-click filming of a moving vehicle, with the action viewable live on the car's interior display as well as recorded for posterity. The flights can also be voice-controlled by the (distracted) driver. The 150,000 Yangwang U8 plug-in hybrid SUV from BYD, the world's largest maker of electric vehicles, sports a DJI drone stored and charged in a dedicated roof space capped with a Thunderbirds-style slide-away panel.

We combine the effectiveness of Reinforcement Learning (RL) and the efficiency of Imitation Learning (IL) in the context of vision-based, autonomous drone racing. We focus on directly processing visual input without explicit state estimation. While RL offers a general framework for learning complex controllers through trial and error, it faces challenges regarding sample efficiency and computational demands due to the high dimensionality of visual inputs. Conversely, IL demonstrates efficiency in learning from visual demonstrations but is limited by the quality of those demonstrations and faces issues like covariate shift. To overcome these limitations, we propose a novel training framework combining RL and IL's advantages. Our framework involves three stages: initial training of a teacher policy using privileged state information, distilling this policy into a student policy using IL, and performance-constrained adaptive RL fine-tuning. Our experiments in both simulated and real-world environments demonstrate that our approach achieves superior performance and robustness than IL or RL alone in navigating a quadrotor through a racing course using only visual information without explicit state estimation.

A critical challenge in deploying unmanned aerial vehicles (UAVs) for autonomous tasks is their ability to navigate in an unknown environment. This paper introduces a novel vision-depth fusion approach for autonomous navigation on nano-UAVs. We combine the visual-based PULP-Dronet convolutional neural network for semantic information extraction, i.e., serving as the global perception, with 8x8px depth maps for close-proximity maneuvers, i.e., the local perception. When tested in-field, our integration strategy highlights the complementary strengths of both visual and depth sensory information. We achieve a 100% success rate over 15 flights in a complex navigation scenario, encompassing straight pathways, static obstacle avoidance, and 90{\deg} turns.

Ukraine launched 35 drones against broad areas of Russia, sparking a brief fire at an oil refinery, targeting Moscow and disrupting electricity in border areas, Russia said early on Sunday, the final day of the country's presidential vote. Moscow accuses Kyiv of election sabotage with its days of strikes on Russian infrastructure, one of the most sweeping air operations on Russian territory since President Vladimir Putin ordered the invasion of neighboring Ukraine two years ago. Putin, all but certain to win re-election, has vowed to punish Ukraine for the attacks.

Motivated by agility, 3D mobility, and low-risk operation compared to human-operated management systems of autonomous unmanned aerial vehicles (UAVs), this work studies UAV-based active wildfire monitoring where a UAV detects fire incidents in remote areas and tracks the fire frontline. A UAV path planning solution is proposed considering realistic wildfire management missions, where a single low-altitude drone with limited power and flight time is available. Noting the limited field of view of commercial low-altitude UAVs, the problem formulates as a partially observable Markov decision process (POMDP), in which wildfire progression outside the field of view causes inaccurate state representation that prevents the UAV from finding the optimal path to track the fire front in limited time. Common deep reinforcement learning (DRL)-based trajectory planning solutions require diverse drone-recorded wildfire data to generalize pre-trained models to real-time systems, which is not currently available at a diverse and standard scale. To narrow down the gap caused by partial observability in the space of possible policies, a belief-based state representation with broad, extensive simulated data is proposed where the beliefs (i.e., ignition probabilities of different grid areas) are updated using a Bayesian framework for the cells within the field of view. The performance of the proposed solution in terms of the ratio of detected fire cells and monitored ignited area (MIA) is evaluated in a complex fire scenario with multiple rapidly growing fire batches, indicating that the belief state representation outperforms the observation state representation both in fire coverage and the distance to fire frontline.

Ukrainian shelling in the southern Russian city of Belgorod has killed two people and a drone attack caused a fire at a Russian oil refinery south of Moscow, officials have said, while Russian authorities claimed to have thwarted a new attempt by saboteurs to cross the border. Saturday's attacks occurred as Russians entered the second day of voting in a presidential election that is all but certain to extend Vladimir Putin's rule by another six years. A man and a woman died in the attack, and three other people were wounded, regional Governor Vyacheslav Gladkov said on the Telegram messaging app. It marks the latest exchange of long-range missiles and rocket fire in the Russia-Ukraine war. Five people were also wounded when a Ukrainian drone hit a car in the village of Glotovo, some two kilometres (1.25 miles) from the Ukrainian border, Gladkov said.

Utilizing unmanned aerial vehicles (UAVs) with edge server to assist terrestrial mobile edge computing (MEC) has attracted tremendous attention. Nevertheless, state-of-the-art schemes based on deterministic optimizations or single-objective reinforcement learning (RL) cannot reduce the backlog of task bits and simultaneously improve energy efficiency in highly dynamic network environments, where the design problem amounts to a sequential decision-making problem. In order to address the aforementioned problems, as well as the curses of dimensionality introduced by the growing number of terrestrial terrestrial users, this paper proposes a distributed multi-objective (MO) dynamic trajectory planning and offloading scheduling scheme, integrated with MORL and the kernel method. The design of n-step return is also applied to average fluctuations in the backlog. Numerical results reveal that the n-step return can benefit the proposed kernel-based approach, achieving significant improvement in the long-term average backlog performance, compared to the conventional 1-step return design. Due to such design and the kernel-based neural network, to which decision-making features can be continuously added, the kernel-based approach can outperform the approach based on fully-connected deep neural network, yielding improvement in energy consumption and the backlog performance, as well as a significant reduction in decision-making and online learning time.

With its army short of ammunition and troops to break the deadlock on the battlefield, Ukraine has increasingly taken the fight behind Russian lines, attacking warships, railways and airfields in an attempt to diminish Moscow's military operations. Most recently, that campaign has focused on oil infrastructure, hitting refineries deep in Russian territory and driving home the country's vulnerability to such attacks. On Tuesday and Wednesday, Ukrainian drones hit four Russian refineries, officials on both sides said, adding to a series of recent attacks that have set fire to depots, fuel tanks and other oil infrastructure across Russia. Since the beginning of the year, Ukraine has claimed responsibility for nearly a dozen such assaults, and local Russian authorities have reported five more.

This paper describes the hardware design and flight demonstration of a small quadrotor with imaging sensors for urban mapping, hazard avoidance, and target tracking research. The vehicle is equipped with five cameras, including two pairs of fisheye stereo cameras that enable a nearly omnidirectional view and a two-axis gimbaled camera. An onboard NVIDIA Jetson Orin Nano computer running the Robot Operating System software is used for data collection. An autonomous tracking behavior was implemented to coordinate the motion of the quadrotor and gimbaled camera to track a moving GPS coordinate. The data collection system was demonstrated through a flight test that tracked a moving GPS-tagged vehicle through a series of roads and parking lots. A map of the environment was reconstructed from the collected images using the Direct Sparse Odometry (DSO) algorithm. The performance of the quadrotor was also characterized by acoustic noise, communication range, battery voltage in hover, and maximum speed tests.

Recently there has been a growing interest in industry and academia, regarding the use of wireless chargers to prolong the operational longevity of unmanned aerial vehicles (commonly knowns as drones). In this paper we consider a charger-assisted drone application: a drone is deployed to observe a set points of interest, while a charger can move to recharge the drone's battery. We focus on the route and charging schedule of the drone and the mobile charger, to obtain high observation utility with the shortest possible time, while ensuring the drone remains operational during task execution. Essentially, this proposed drone-charger scheduling problem is a multi-stage decision-making process, in which the drone and the mobile charger act as two agents who cooperate to finish a task. The discrete-continuous hybrid action space of the two agents poses a significant challenge in our problem. To address this issue, we present a hybrid-action deep reinforcement learning framework, called HaDMC, which uses a standard policy learning algorithm to generate latent continuous actions. Motivated by representation learning, we specifically design and train an action decoder. It involves two pipelines to convert the latent continuous actions into original discrete and continuous actions, by which the drone and the charger can directly interact with environment. We embed a mutual learning scheme in model training, emphasizing the collaborative rather than individual actions. We conduct extensive numerical experiments to evaluate HaDMC and compare it with state-of-the-art deep reinforcement learning approaches. The experimental results show the effectiveness and efficiency of our solution.