Multi-Agent Path Finding (MAPF) in crowded environments presents a challenging problem in motion planning, aiming to find collision-free paths for all agents in the system. MAPF finds a wide range of applications in various domains, including aerial swarms, autonomous warehouse robotics, and self-driving vehicles. The current approaches for MAPF can be broadly categorized into two main categories: centralized and decentralized planning. Centralized planning suffers from the curse of dimensionality and thus does not scale well in large and complex environments. On the other hand, decentralized planning enables agents to engage in real-time path planning within a partially observable environment, demonstrating implicit coordination. However, they suffer from slow convergence and performance degradation in dense environments. In this paper, we introduce CRAMP, a crowd-aware decentralized approach to address this problem by leveraging reinforcement learning guided by a boosted curriculum-based training strategy. We test CRAMP on simulated environments and demonstrate that our method outperforms the state-of-the-art decentralized methods for MAPF on various metrics. CRAMP improves the solution quality up to 58% measured in makespan and collision count, and up to 5% in success rate in comparison to previous methods.

Controller design for soft robots is challenging due to nonlinear deformation and high degrees of freedom of flexible material. The data-driven approach is a promising solution to the controller design problem for soft robots. However, the existing data-driven controller design methods for soft robots suffer from two drawbacks: (i) they require excessively long training time, and (ii) they may result in potentially inefficient controllers. This paper addresses these issues by developing two memory-based controllers for soft robots that can be trained in a data-driven fashion: the finite memory controller (FMC) approach and the long short-term memory (LSTM) based approach. An FMC stores the tracking errors at different time instances and computes the actuation signal according to a weighted sum of the stored tracking errors. We develop three reinforcement learning algorithms for computing the optimal weights of an FMC using the Q-learning, soft actor-critic, and deterministic policy gradient (DDPG) methods. An LSTM-based controller is composed of an LSTM network where the inputs of the network are the robot's desired configuration and current configuration. The LSTM network computes the required actuation signal for the soft robot to follow the desired configuration. We study the performance of the proposed approaches in controlling a soft finger where, as benchmarks, we use the existing reinforcement learning (RL) based controllers and proportional-integral-derivative (PID) controllers. Our numerical results show that the training time of the proposed memory-based controllers is significantly shorter than that of the classical RL-based controllers. Moreover, the proposed controllers achieve a smaller tracking error compared with the classical RL algorithms and the PID controller.

Y ou can use this material personally. Abstract -- Nowadays, as the need for capacity continues to grow, entirely novel services are emerging. A solid cloud-network integrated infrastructure is necessary to supply these services in a real-time responsive, and scalable way. Due to their diverse characteristics and limited capacity, communication and computing resources must be collaboratively managed to unleash their full potential. Although several innovative methods have been proposed to orchestrate the resources, most ignored network resources or relaxed the network as a simple graph, focusing only on cloud resources. This paper fills the gap by studying the joint problem of communication and computing resource allocation, dubbed CCRA, including function placement and assignment, traffic prioritization, and path selection considering capacity constraints and quality requirements, to minimize total cost. We formulate the problem as a non-linear programming model and propose two approaches, dubbed B&B-CCRA and WF-CCRA, based on the Branch & Bound and Water-Filling algorithms to solve it when the system is fully known. Then, for partially known systems, a Double Deep Q-Learning (DDQL) architecture is designed. Numerical simulations show that B&B-CCRA optimally solves the problem, whereas WF-CCRA delivers near-optimal solutions in a substantially shorter time. Furthermore, it is demonstrated that DDQL-CCRA obtains near-optimal solutions in the absence of request-specific information. Nowadays, an increase in data flow has resulted in a 1000-fold increase in network capacity, which is the primary driver of network evolution. While this demand for capacity will continue to grow, the Internet of Everything is forging a paradigm shift to new-born perceptions, bringing a range of novel services with rigorous deterministic criteria, such as connected robotics, smart healthcare, autonomous transportation, and extended reality [1]. These services will be provisioned by establishing functional components, Virtual Network Functions (VNFs), which will generate and consume vast amounts of data that must be processed in real-time to ensure service responsiveness and scalability. In these circumstances, a distributed cloud architecture is essential [2], which could be implemented via a solid cloud-network integrated infrastructure built of distinct domains in Beyond 5G (B5G) [3]. These domains can be distinguished by the technology employed, including radio access, transport, and core networks, as well as edge, access, aggregation, regional, and central clouds. Moreover, these resources can be virtualized using technologies such as Network Function Vir-tualization (NFV), which enables the construction of separate virtual entities on top of this physical infrastructure [4], [5].

The Metaverse is a new paradigm that aims to create a virtual environment consisting of numerous worlds, each of which will offer a different set of services. To deal with such a dynamic and complex scenario, considering the stringent quality of service requirements aimed at the 6th generation of communication systems (6G), one potential approach is to adopt self-sustaining strategies, which can be realized by employing Adaptive Artificial Intelligence (Adaptive AI) where models are continually re-trained with new data and conditions. One aspect of self-sustainability is the management of multiple access to the frequency spectrum. Although several innovative methods have been proposed to address this challenge, mostly using Deep Reinforcement Learning (DRL), the problem of adapting agents to a non-stationary environment has not yet been precisely addressed. This paper fills in the gap in the current literature by investigating the problem of multiple access in multi-channel environments to maximize the throughput of the intelligent agent when the number of active User Equipments (UEs) may fluctuate over time. To solve the problem, a Double Deep Q-Learning (DDQL) technique empowered by Continual Learning (CL) is proposed to overcome the non-stationary situation, while the environment is unknown. Numerical simulations demonstrate that, compared to other well-known methods, the CL-DDQL algorithm achieves significantly higher throughputs with a considerably shorter convergence time in highly dynamic scenarios.

Learning from human demonstrations (behavior cloning) is a cornerstone of robot learning. However, most behavior cloning algorithms require a large number of demonstrations to learn a task, especially for general tasks that have a large variety of initial conditions. Humans, however, can learn to complete tasks, even complex ones, after only seeing one or two demonstrations. Our work seeks to emulate this ability, using behavior cloning to learn a task given only a single human demonstration. We achieve this goal by using linear transforms to augment the single demonstration, generating a set of trajectories for a wide range of initial conditions. With these demonstrations, we are able to train a behavior cloning agent to successfully complete three block manipulation tasks. Additionally, we developed a novel addition to the temporal ensembling method used by action chunking agents during inference. By incorporating the standard deviation of the action predictions into the ensembling method, our approach is more robust to unforeseen changes in the environment, resulting in significant performance improvements.

Decentralized exchanges (DEXs) are a cornerstone of decentralized finance (DeFi), allowing users to trade cryptocurrencies without the need for third-party authorization. Investors are incentivized to deposit assets into liquidity pools, against which users can trade directly, while paying fees to liquidity providers (LPs). However, a number of unresolved issues related to capital efficiency and market risk hinder DeFi's further development. Uniswap V3, a leading and groundbreaking DEX project, addresses capital efficiency by enabling LPs to concentrate their liquidity within specific price ranges for deposited assets. Nevertheless, this approach exacerbates market risk, as LPs earn trading fees only when asset prices are within these predetermined brackets. To mitigate this issue, this paper introduces a deep reinforcement learning (DRL) solution designed to adaptively adjust these price ranges, maximizing profits and mitigating market risks. Our approach also neutralizes price-change risks by hedging the liquidity position through a rebalancing portfolio in a centralized futures exchange. The DRL policy aims to optimize trading fees earned by LPs against associated costs, such as gas fees and hedging expenses, which is referred to as loss-versus-rebalancing (LVR). Using simulations with a profit-and-loss (PnL) benchmark, our method demonstrates superior performance in ETH/USDC and ETH/USDT pools compared to existing baselines. We believe that this strategy not only offers investors a valuable asset management tool but also introduces a new incentive mechanism for DEX designers.

Unmanned aerial vehicles (UAVs) are capable of surveying expansive areas, but their operational range is constrained by limited battery capacity. The deployment of mobile recharging stations using unmanned ground vehicles (UGVs) significantly extends the endurance and effectiveness of UAVs. However, optimizing the routes of both UAVs and UGVs, known as the UAV-UGV cooperative routing problem, poses substantial challenges, particularly with respect to the selection of recharging locations. Here in this paper, we leverage reinforcement learning (RL) for the purpose of identifying optimal recharging locations while employing constraint programming to determine cooperative routes for the UAV and UGV. Our proposed framework is then benchmarked against a baseline solution that employs Genetic Algorithms (GA) to select rendezvous points. Our findings reveal that RL surpasses GA in terms of reducing overall mission time, minimizing UAV-UGV idle time, and mitigating energy consumption for both the UAV and UGV. These results underscore the efficacy of incorporating heuristics to assist RL, a method we refer to as heuristics-assisted RL, in generating high-quality solutions for intricate routing problems.

Traffic congestion in dense urban centers presents an economical and environmental burden. In recent years, the availability of vehicle-to-anything communication allows for the transmission of detailed vehicle states to the infrastructure that can be used for intelligent traffic light control. The other way around, the infrastructure can provide vehicles with advice on driving behavior, such as appropriate velocities, which can improve the efficacy of the traffic system. Several research works applied deep reinforcement learning to either traffic light control or vehicle speed advice. In this work, we propose a first attempt to jointly learn the control of both. We show this to improve the efficacy of traffic systems. In our experiments, the joint control approach reduces average vehicle trip delays, w.r.t. controlling only traffic lights, in eight out of eleven benchmark scenarios. Analyzing the qualitative behavior of the vehicle speed advice policy, we observe that this is achieved by smoothing out the velocity profile of vehicles nearby a traffic light. Learning joint control of traffic signaling and speed advice in the real world could help to reduce congestion and mitigate the economical and environmental repercussions of today's traffic systems.

Cognitive maps play a crucial role in facilitating flexible behaviour by representing spatial and conceptual relationships within an environment. The ability to learn and infer the underlying structure of the environment is crucial for effective exploration and navigation. This paper introduces a hierarchical active inference model addressing the challenge of inferring structure in the world from pixel-based observations. We propose a three-layer hierarchical model consisting of a cognitive map, an allocentric, and an egocentric world model, combining curiosity-driven exploration with goal-oriented behaviour at the different levels of reasoning from context to place to motion. This allows for efficient exploration and goal-directed search in room-structured mini-grid environments.

Reinforcement Learning (RL) has the potential to surpass human performance in driving without needing any expert supervision. Despite its promise, the state-of-the-art in sensorimotor self-driving is dominated by imitation learning methods due to the inherent shortcomings of RL algorithms. Nonetheless, RL agents are able to discover highly successful policies when provided with privileged ground truth representations of the environment. In this work, we investigate what separates privileged RL agents from sensorimotor agents for urban driving in order to bridge the gap between the two. We propose vision-based deep learning models to approximate the privileged representations from sensor data. In particular, we identify aspects of state representation that are crucial for the success of the RL agent such as desired route generation and stop zone prediction, and propose solutions to gradually develop less privileged RL agents. We also observe that bird's-eye-view models trained on offline datasets do not generalize to online RL training due to distribution mismatch. Through rigorous evaluation on the CARLA simulation environment, we shed light on the significance of the state representations in RL for autonomous driving and point to unresolved challenges for future research.