[1] Liu, Chuhao and Shaojie Shen. “An Augmented Reality Interaction Interface for Autonomous Drone.” 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (2020): 11419-11424.
[2] Anwar, Malik Aqeel and Arijit Raychowdhury. “NavREn-Rl: Learning to fly in real environment via end-to-end deep reinforcement learning using monocular images.” 2018 25th International Conference on Mechatronics and Machine Vision in Practice (M2VIP) (2018): 1-6.
[3] Karmanova, Ekaterina et al. “SwarmPlay: Interactive Tic-tac-toe Board Game with Swarm of Nano-UAVs driven by Reinforcement Learning.” 2021 30th IEEE International Conference on Robot & Human Interactive Communication (RO-MAN) (2021): 1269-1274.
, as listed below. (2)
phrase
, etc. (3)
phrase
Example: There are a variety of home robot assistants, e.g., Kury, Buddy, Aido, etc.
..., e.g., ... (3)
phrase
Example: There are a variety of home robot assistants, e.g., Kury, Buddy, Aido, etc.
a fresh perspective on (3)
phrase
accomplish (2)
verb
achieve or complete successfully
Example: The agent accomplishes this by taking the opposite actions.
en: выполнять
adjustable (3)
adjective
capable of being adjusted: adjust - alter or move (something) slightly in order to achieve the desired fit, appearance, or result.
Example: One of the potential domains for the human-robot interaction research is physical board games with an adjustable structure level.
en: регулируемый
aggregation (2)
noun
the formation of a number of things into a cluster.
Example: Collection of small number of expert data and knowledge based data aggregation is integrated into the RL process to aid convergence.
en: сбор
an end-to-end methodology (2)
As opposed to (2)
phrase
as proven by the results (1)
phrase
carry out (2)
phrase
perform a task.
Example: Experimentation is carried out on a Parrot AR drone in different indoor arenas and the results are compared with other baseline technologies.
en: исполнять
clipping (2)
noun
process of cutting
Example: Clipping Q values in Bellman equation
en: отсечение
convergence (2)
noun
the act of converging and especially moving toward union or uniformity
Example: The basic RL algorithm often suffers from limited convergence, which mainly emerges because the Bellman equation tends to over-estimate the Q-values due to its non-linear nature.
en: сходимость
Despite the limitations (1)
phrase
diverge (2)
verb
(of a series) increase indefinitely as more of its terms are added.
Example: This ensures that the Q-value updates do not diverge.
en: расходиться
drag (1)
verb
pull (someone or something) along forcefully, roughly, or with difficulty.
Example: Then, the virtual drone can be selected by hand gesture and be dragged to the spatial target position.
en: тянуть
Due to its superior test accuracy, (2)
phrase
dynamic windowing
phrase
an online collision avoidance strategy for mobile robots
Example: This global depth based dynamic windowing can be seen in Figure 2.
encounter (3)
verb
unexpectedly be faced with or experience (something hostile or difficult).
Example: Temporal Difference (TD) learning is one of the most frequently encountered model-free learning algorithms in RL as it enables the agent to learn through every single action it takes.
* The word is used incorrectly in the article.
en: сталкиваться
establish (3)
verb
set up on a firm or permanent basis.
Example: Swarm then will establish the formations and communication nodes to achieve safe data transferring.
en: устанавливать
Figure 4 shows the comparison of (2)
fine-tuning (1)
phrase
make small adjustments to (something) in order to achieve the best or a desired performance
Example: The neural network trained in the simulations is then also tested in the real environment without any fine-tuning and has shown to perform well.
en: тонкая настройка
Hence (2)
adverb
as a consequence; for this reason.
Example: Hence extra sensing capabilities need to be added to the agent giving
it a notion of depth which not only adds to the computation cost but also to the weight of the agent.
en: следовательно
In most of the cases (1)
In our approach, (2)
incorporate (1)
verb
take in or contain (something) as part of a whole; include.
Example: Human drone interaction in autonomous navigation incorporates spatial interaction tasks, including reconstructed 3D map from the drone and human desired target
position.
en: включать
intervention (1)
the action or process of intervening
Example: All of these previous demonstrations and approaches, in spite of their many successes, either require considerable human/expert intervention, handcrafted algorithms or are implemented offline in simulations, where the simulated and the real endowments need to be nearly identical.
en: вмешательство
Irrespective (2)
adverb
regardless or in spite
Example: Irrespective of the application, navigating autonomously is one of the key desirable features of UAVs both indoors and outdoors.
en: независимо
maintain (1)
verb
provide with necessities for life or existence.
Example: Surfel Fusion maintains a series of surfel maps with attached pose graphs to ensure global consistency.
en: поддерживать
narrow (2)
adjective
of small width in relation to length.
Example: If the environment is expected to have narrow (wide) obstacles, the value of n is relatively smaller (larger).
en: узкий
occupancy map (1)
phrase
Occupancy Grid Mapping refers to a family of computer algorithms in probabilistic robotics for mobile robots which address the problem of generating maps from noisy and uncertain sensor measurement data, with the assumption that the robot pose is known.
Example: The occupancy map uses cubic volumetric elements (voxels) to represent the 3D structure around the drone.
en: Карта занятости
One of the potential domains for (3)
Our principal goal is to (1)
Over the past decade, (2)
phrase
persistent (2)
adjective
continuing firmly or obstinately in an opinion or course of action in spite of difficulty or opposition.
Example: Making expert data a persistent part of the experience replay helps avoid the neural network from forgetting what it had learned in the pre-training phase.
en: постоянный
presume (3)
verb
suppose that something is the case on the basis of probability.
Example: Most game theory algorithms applied in human-robot interaction scenarios presume an opponent (human) and an agent (robot) to be ideal.
en: предполагать
Real-time experimentation (2)
phrase
Example: Real-time experimentation are carried out to validate the proposed approach for drone navigation.
reconnaissance (2)
noun
military observation of a region to locate an enemy or ascertain strategic features.
Example: Over the past decade, there has been considerable success in using Unmanned Aerial Vehicles (UAVs) or drones in varied applications such as reconnaissance, surveying, rescuing and mapping.
en: разведка
replay memory (2)
phrase
Experience replay memory in reinforcement learning enables agents to remember and reuse past experiences.
Example: A mini-batch of size n is randomly sampled from the replay memory D and used to minimize the loss defined in equation 5 through gradient descent.
en: воспроизводимая память
Restrict (2)
verb
put a limit on, keep under control
Example: Restricting the range of rewards
en: ограничивать
robust (1)
adjective
(of a system, organization, etc.) able to withstand or overcome adverse conditions.
Example: It provides a very robust and accurate VIO.
en: надежный
spatial (1)
adjective
relating to or occupying space
Example: In spatial interaction tasks, reality devices have obvious advantages in rendering and conveying human intention.
en: пространственный
swarm intelligence (3)
phrase
the collective behavior of decentralized, self-organized systems, natural or artificial.
Example: SwarmPlay, where each playing component is presented by an individual drone that has its own mobility and swarm intelligence to win against a human player.
en: интеллект роя
tend (2)
verb
regularly or frequently behave in a particular way or have a certain characteristic
Example: The basic RL algorithm often suffers from limited convergence, which mainly emerges because the Bellman equation tends to over-estimate the Q-values due to its non-linear nature.
en: иметь тенденцию
The amount of data required for (1)
the function is designed keeping in mind the cost (2)
phrase
the goal achieved through (2)
phrase
the learning phase (2)
phrase
The main contributions of the paper are as follows: (1)
The preliminary user study revealed that (3)
phrase
This work was supported (1)
time instant (2)
phrase
a single, atomic time point in the time domain.
Example: At each time instant, the drone observes the current camera frame s.
en: момент времени
to find a correlation between (3)
To our knowledge, (3)
phrase
to statistically verify (1)
phrase
To sum up, (1)
phrase
We assess the performance of ( ) by comparing it against the ( ) (2)