This project has conducted research on robot path planning based on Visual SLAM. The main work of this project is as follows: (1) Construction of Visual SLAM system. Research has been conducted on the basic architecture of Visual SLAM. A Visual SLAM system is developed based on ORB-SLAM3 system, which can conduct dense point cloud mapping. (2) The map suitable for two-dimensional path planning is obtained through map conversion. This part converts the dense point cloud map obtained by Visual SLAM system into an octomap and then performs projection transformation to the grid map. The map conversion converts the dense point cloud map containing a large amount of redundant map information into an extremely lightweight grid map suitable for path planning. (3) Research on path planning algorithm based on reinforcement learning. This project has conducted experimental comparisons between the Q-learning algorithm, the DQN algorithm, and the SARSA algorithm, and found that DQN is the algorithm with the fastest convergence and best performance in high-dimensional complex environments. This project has conducted experimental verification of the Visual SLAM system in a simulation environment. The experimental results obtained based on open-source dataset and self-made dataset prove the feasibility and effectiveness of the designed Visual SLAM system. At the same time, this project has also conducted comparative experiments on the three reinforcement learning algorithms under the same experimental condition to obtain the optimal algorithm under the experimental condition.
This paper studies continuous-time risk-sensitive reinforcement learning (RL) under the entropy-regularized, exploratory diffusion process formulation with the exponential-form objective. The risk-sensitive objective arises either as the agent's risk attitude or as a distributionally robust approach against the model uncertainty. Owing to the martingale perspective in Jia and Zhou (2023) the risk-sensitive RL problem is shown to be equivalent to ensuring the martingale property of a process involving both the value function and the q-function, augmented by an additional penalty term: the quadratic variation of the value process, capturing the variability of the value-to-go along the trajectory. This characterization allows for the straightforward adaptation of existing RL algorithms developed for non-risk-sensitive scenarios to incorporate risk sensitivity by adding the realized variance of the value process. Additionally, I highlight that the conventional policy gradient representation is inadequate for risk-sensitive problems due to the nonlinear nature of quadratic variation; however, q-learning offers a solution and extends to infinite horizon settings. Finally, I prove the convergence of the proposed algorithm for Merton's investment problem and quantify the impact of temperature parameter on the behavior of the learning procedure. I also conduct simulation experiments to demonstrate how risk-sensitive RL improves the finite-sample performance in the linear-quadratic control problem.
In this work, we propose a new setting of continual learning: data-incremental continual offline reinforcement learning (DICORL), in which an agent is asked to learn a sequence of datasets of a single offline reinforcement learning (RL) task continually, instead of learning a sequence of offline RL tasks with respective datasets. Then, we propose that this new setting will introduce a unique challenge to continual learning: active forgetting, which means that the agent will forget the learnt skill actively. The main reason for active forgetting is conservative learning used by offline RL, which is used to solve the overestimation problem. With conservative learning, the offline RL method will suppress the value of all actions, learnt or not, without selection, unless it is in the just learning dataset. Therefore, inferior data may overlay premium data because of the learning sequence. To solve this problem, we propose a new algorithm, called experience-replay-based ensemble implicit Q-learning (EREIQL), which introduces multiple value networks to reduce the initial value and avoid using conservative learning, and the experience replay to relieve catastrophic forgetting. Our experiments show that EREIQL relieves active forgetting in DICORL and performs well.
Reinforcement Learning From Human Feedback (RLHF) has been critical to the success of the latest generation of generative AI models. In response to the complex nature of the classical RLHF pipeline, direct alignment algorithms such as Direct Preference Optimization (DPO) have emerged as an alternative approach. Although DPO solves the same objective as the standard RLHF setup, there is a mismatch between the two approaches. Standard RLHF deploys reinforcement learning in a specific token-level MDP, while DPO is derived as a bandit problem in which the whole response of the model is treated as a single arm. In this work we rectify this difference. We theoretically show that we can derive DPO in the token-level MDP as a general inverse Q-learning algorithm, which satisfies the Bellman equation. Using our theoretical results, we provide three concrete empirical insights. First, we show that because of its token level interpretation, DPO is able to perform some type of credit assignment. Next, we prove that under the token level formulation, classical search-based algorithms, such as MCTS, which have recently been applied to the language generation space, are equivalent to likelihood-based search on a DPO policy. Empirically we show that a simple beam search yields meaningful improvement over the base DPO policy. Finally, we show how the choice of reference policy causes implicit rewards to decline during training. We conclude by discussing applications of our work, including information elicitation in multi-turn dialogue, reasoning, agentic applications and end-to-end training of multi-model systems.
Reinforcement learning has recently gained unprecedented popularity, yet it still grapples with sample inefficiency. Addressing this challenge, federated reinforcement learning (FedRL) has emerged, wherein agents collaboratively learn a single policy by aggregating local estimations. However, this aggregation step incurs significant communication costs. In this paper, we propose CompFedRL, a communication-efficient FedRL approach incorporating both \textit{periodic aggregation} and (direct/error-feedback) compression mechanisms. Specifically, we consider compressed federated $Q$-learning with a generative model setup, where a central server learns an optimal $Q$-function by periodically aggregating compressed $Q$-estimates from local agents. For the first time, we characterize the impact of these two mechanisms (which have remained elusive) by providing a finite-time analysis of our algorithm, demonstrating strong convergence behaviors when utilizing either direct or error-feedback compression. Our bounds indicate improved solution accuracy concerning the number of agents and other federated hyperparameters while simultaneously reducing communication costs. To corroborate our theory, we also conduct in-depth numerical experiments to verify our findings, considering Top-$K$ and Sparsified-$K$ sparsification operators.
Embodied visual tracking is to follow a target object in dynamic 3D environments using an agent's egocentric vision. This is a vital and challenging skill for embodied agents. However, existing methods suffer from inefficient training and poor generalization. In this paper, we propose a novel framework that combines visual foundation models(VFM) and offline reinforcement learning(offline RL) to empower embodied visual tracking. We use a pre-trained VFM, such as "Tracking Anything", to extract semantic segmentation masks with text prompts. We then train a recurrent policy network with offline RL, e.g., Conservative Q-Learning, to learn from the collected demonstrations without online interactions. To further improve the robustness and generalization of the policy network, we also introduce a mask re-targeting mechanism and a multi-level data collection strategy. In this way, we can train a robust policy within an hour on a consumer-level GPU, e.g., Nvidia RTX 3090. We evaluate our agent on several high-fidelity environments with challenging situations, such as distraction and occlusion. The results show that our agent outperforms state-of-the-art methods in terms of sample efficiency, robustness to distractors, and generalization to unseen scenarios and targets. We also demonstrate the transferability of the learned agent from virtual environments to a real-world robot.
Over the past decades, the increase in both frequency and intensity of large-scale wildfires due to climate change has emerged as a significant natural threat. The pressing need to design resilient landscapes capable of withstanding such disasters has become paramount, requiring the development of advanced decision-support tools. Existing methodologies, including Mixed Integer Programming, Stochastic Optimization, and Network Theory, have proven effective but are hindered by computational demands, limiting their applicability. In response to this challenge, we propose using artificial intelligence techniques, specifically Deep Reinforcement Learning, to address the complex problem of firebreak placement in the landscape. We employ value-function based approaches like Deep Q-Learning, Double Deep Q-Learning, and Dueling Double Deep Q-Learning. Utilizing the Cell2Fire fire spread simulator combined with Convolutional Neural Networks, we have successfully implemented a computational agent capable of learning firebreak locations within a forest environment, achieving good results. Furthermore, we incorporate a pre-training loop, initially teaching our agent to mimic a heuristic-based algorithm and observe that it consistently exceeds the performance of these solutions. Our findings underscore the immense potential of Deep Reinforcement Learning for operational research challenges, especially in fire prevention. Our approach demonstrates convergence with highly favorable results in problem instances as large as 40 x 40 cells, marking a significant milestone in applying Reinforcement Learning to this critical issue. To the best of our knowledge, this study represents a pioneering effort in using Reinforcement Learning to address the aforementioned problem, offering promising perspectives in fire prevention and landscape management
Arterial traffic interacts with freeway traffic, yet the two are controlled independently. Arterial traffic signals do not take into account freeway traffic and how ramps control ingress traffic and have no control over egress traffic from the freeway. This often results in long queues in either direction that block ramps and spill over to arterial streets or freeway lanes. In this paper, we propose an adaptive arterial traffic control strategy that combines traffic signal control (TSC) and dynamic speed offset (DSO) coordination using a Q-learning algorithm for a traffic network that involves a freeway segment and adjacent arterial streets. The TSC agent computes the signal cycle length and split based on observed intersection demands and adjacent freeway off-ramp queues. The DSO agent computes the relative offset and the recommended speeds of both ways between consecutive intersections based on their physical distance, intersection queues, and signal cycles. We evaluate the performance of the proposed arterial traffic control strategy using microscopic traffic simulations of an arterial corridor with seven intersections near the I-710 freeway. The proposed QL-based control significantly outperforms a fixed-time control and MAXBAND in terms of the travel time and the number of stops under low or moderate demands. In high-demand scenarios, the travel-time benefit provided by the QL-based control is reduced as it mitigates off-ramp and intersection queues, which is a necessary trade-off in our perspective. In addition, mutual benefit is obtained by implementing freeway and arterial traffic control simultaneously.
This work describes a technique for active rejection of multiple independent and time-correlated stochastic disturbances for a nonlinear flexible inverted pendulum with cart system with uncertain model parameters. The control law is determined through deep reinforcement learning, specifically with a continuous actor-critic variant of deep Q-learning known as Deep Deterministic Policy Gradient, while the disturbance magnitudes evolve via independent stochastic processes. Simulation results are then compared with those from a classical control system.
Motivated by Q-learning, we study nonsmooth contractive stochastic approximation (SA) with constant stepsize. We focus on two important classes of dynamics: 1) nonsmooth contractive SA with additive noise, and 2) synchronous and asynchronous Q-learning, which features both additive and multiplicative noise. For both dynamics, we establish weak convergence of the iterates to a stationary limit distribution in Wasserstein distance. Furthermore, we propose a prelimit coupling technique for establishing steady-state convergence and characterize the limit of the stationary distribution as the stepsize goes to zero. Using this result, we derive that the asymptotic bias of nonsmooth SA is proportional to the square root of the stepsize, which stands in sharp contrast to smooth SA. This bias characterization allows for the use of Richardson-Romberg extrapolation for bias reduction in nonsmooth SA.
The target set selection problem (TSS) asks for a set of vertices such that an influence spreading process started in these vertices reaches the whole graph. The current state of the art for this NP-hard problem are three recently proposed randomized search heuristics, namely a biased random-key genetic algorithm (BRKGA) obtained from extensive parameter tuning, a max-min ant system (MMAS), and a MMAS using Q-learning with a graph convolutional network. We show that the BRKGA with two simple modifications and without the costly parameter tuning obtains significantly better results. Our first modification is to simply choose all parameters of the BRKGA in each iteration randomly from a power-law distribution. The resulting parameterless BRKGA is already competitive with the tuned BRKGA, as our experiments on the previously used benchmarks show. We then add a natural greedy heuristic, namely to repeatedly discard small-degree vertices that are not necessary for reaching the whole graph. The resulting algorithm consistently outperforms all of the state-of-the-art algorithms. Besides providing a superior algorithm for the TSS problem, this work shows that randomized parameter choices and elementary greedy heuristics can give better results than complex algorithms and costly parameter tuning.
Recent reinforcement learning approaches have shown surprisingly strong capabilities of bang-bang policies for solving continuous control benchmarks. The underlying coarse action space discretizations often yield favourable exploration characteristics while final performance does not visibly suffer in the absence of action penalization in line with optimal control theory. In robotics applications, smooth control signals are commonly preferred to reduce system wear and energy efficiency, but action costs can be detrimental to exploration during early training. In this work, we aim to bridge this performance gap by growing discrete action spaces from coarse to fine control resolution, taking advantage of recent results in decoupled Q-learning to scale our approach to high-dimensional action spaces up to dim(A) = 38. Our work indicates that an adaptive control resolution in combination with value decomposition yields simple critic-only algorithms that yield surprisingly strong performance on continuous control tasks.
We introduce the Laser Learning Environment (LLE), a collaborative multi-agent reinforcement learning environment in which coordination is central. In LLE, agents depend on each other to make progress (interdependence), must jointly take specific sequences of actions to succeed (perfect coordination), and accomplishing those joint actions does not yield any intermediate reward (zero-incentive dynamics). The challenge of such problems lies in the difficulty of escaping state space bottlenecks caused by interdependence steps since escaping those bottlenecks is not rewarded. We test multiple state-of-the-art value-based MARL algorithms against LLE and show that they consistently fail at the collaborative task because of their inability to escape state space bottlenecks, even though they successfully achieve perfect coordination. We show that Q-learning extensions such as prioritized experience replay and n-steps return hinder exploration in environments with zero-incentive dynamics, and find that intrinsic curiosity with random network distillation is not sufficient to escape those bottlenecks. We demonstrate the need for novel methods to solve this problem and the relevance of LLE as cooperative MARL benchmark.
Accounting for the uncertainty of value functions boosts exploration in Reinforcement Learning (RL). Our work introduces Maximum Mean Discrepancy Q-Learning (MMD-QL) to improve Wasserstein Q-Learning (WQL) for uncertainty propagation during Temporal Difference (TD) updates. MMD-QL uses the MMD barycenter for this purpose, as MMD provides a tighter estimate of closeness between probability measures than the Wasserstein distance. Firstly, we establish that MMD-QL is Probably Approximately Correct in MDP (PAC-MDP) under the average loss metric. Concerning the accumulated rewards, experiments on tabular environments show that MMD-QL outperforms WQL and other algorithms. Secondly, we incorporate deep networks into MMD-QL to create MMD Q-Network (MMD-QN). Making reasonable assumptions, we analyze the convergence rates of MMD-QN using function approximation. Empirical results on challenging Atari games demonstrate that MMD-QN performs well compared to benchmark deep RL algorithms, highlighting its effectiveness in handling large state-action spaces.
Autonomous robots operating in complex environments face the critical challenge of identifying and utilizing environmental cover for covert navigation to minimize exposure to potential threats. We propose EnCoMP, an enhanced navigation framework that integrates offline reinforcement learning and our novel Adaptive Threat-Aware Visibility Estimation (ATAVE) algorithm to enable robots to navigate covertly and efficiently in diverse outdoor settings. ATAVE is a dynamic probabilistic threat modeling technique that we designed to continuously assess and mitigate potential threats in real-time, enhancing the robot's ability to navigate covertly by adapting to evolving environmental and threat conditions. Moreover, our approach generates high-fidelity multi-map representations, including cover maps, potential threat maps, height maps, and goal maps from LiDAR point clouds, providing a comprehensive understanding of the environment. These multi-maps offer detailed environmental insights, helping in strategic navigation decisions. The goal map encodes the relative distance and direction to the target location, guiding the robot's navigation. We train a Conservative Q-Learning (CQL) model on a large-scale dataset collected from real-world environments, learning a robust policy that maximizes cover utilization, minimizes threat exposure, and maintains efficient navigation. We demonstrate our method's capabilities on a physical Jackal robot, showing extensive experiments across diverse terrains. These experiments demonstrate EnCoMP's superior performance compared to state-of-the-art methods, achieving a 95% success rate, 85% cover utilization, and reducing threat exposure to 10.5%, while significantly outperforming baselines in navigation efficiency and robustness.
Reinforcement learning (RL) algorithms have become indispensable tools in artificial intelligence, empowering agents to acquire optimal decision-making policies through interactions with their environment and feedback mechanisms. This study explores the performance of RL agents in both two-dimensional (2D) and three-dimensional (3D) environments, aiming to research the dynamics of learning across different spatial dimensions. A key aspect of this investigation is the absence of pre-made libraries for learning, with the algorithm developed exclusively through computational mathematics. The methodological framework centers on RL principles, employing a Q-learning agent class and distinct environment classes tailored to each spatial dimension. The research aims to address the question: How do reinforcement learning agents adapt and perform in environments of varying spatial dimensions, particularly in 2D and 3D settings? Through empirical analysis, the study evaluates agents' learning trajectories and adaptation processes, revealing insights into the efficacy of RL algorithms in navigating complex, multi-dimensional spaces. Reflections on the findings prompt considerations for future research, particularly in understanding the dynamics of learning in higher-dimensional environments.
We consider a setting in which $N$ agents aim to speedup a common Stochastic Approximation (SA) problem by acting in parallel and communicating with a central server. We assume that the up-link transmissions to the server are subject to asynchronous and potentially unbounded time-varying delays. To mitigate the effect of delays and stragglers while reaping the benefits of distributed computation, we propose \texttt{DASA}, a Delay-Adaptive algorithm for multi-agent Stochastic Approximation. We provide a finite-time analysis of \texttt{DASA} assuming that the agents' stochastic observation processes are independent Markov chains. Significantly advancing existing results, \texttt{DASA} is the first algorithm whose convergence rate depends only on the mixing time $\tau_{mix}$ and on the average delay $\tau_{avg}$ while jointly achieving an $N$-fold convergence speedup under Markovian sampling. Our work is relevant for various SA applications, including multi-agent and distributed temporal difference (TD) learning, Q-learning and stochastic optimization with correlated data.
This paper studies the remote estimation of multiple Markov sources over a lossy and rate-constrained channel. Unlike most existing studies that treat all source states equally, we exploit the \emph{semantics of information} and consider that the remote actuator has different tolerances for the estimation errors. We aim to find an optimal scheduling policy that minimizes the long-term \textit{state-dependent} costs of estimation errors under a transmission frequency constraint. The optimal scheduling problem is formulated as a \emph{constrained Markov decision process} (CMDP). We show that the optimal Lagrangian cost follows a piece-wise linear and concave (PWLC) function, and the optimal policy is, at most, a randomized mixture of two simple deterministic policies. By exploiting the structural results, we develop a new \textit{intersection search} algorithm that finds the optimal policy using only a few iterations. We further propose a reinforcement learning (RL) algorithm to compute the optimal policy without knowing \textit{a priori} the channel and source statistics. To avoid the ``curse of dimensionality" in MDPs, we propose an online low-complexity \textit{drift-plus-penalty} (DPP) algorithm. Numerical results show that continuous transmission is inefficient, and remarkably, our semantic-aware policies can attain the optimum by strategically utilizing fewer transmissions by exploiting the timing of the important information.
As Machine Learning grows in popularity across various fields, equity has become a key focus for the AI community. However, fairness-oriented approaches are still underexplored in smart mobility. Addressing this gap, our study investigates the balance between performance optimization and algorithmic fairness in shared micromobility services providing a novel framework based on Reinforcement Learning. Exploiting Q-learning, the proposed methodology achieves equitable outcomes in terms of the Gini index across different areas characterized by their distance from central hubs. Through vehicle rebalancing, the provided scheme maximizes operator performance while ensuring fairness principles for users, reducing iniquity by up to 85% while only increasing costs by 30% (w.r.t. applying no equity adjustment). A case study with synthetic data validates our insights and highlights the importance of fairness in urban micromobility (source code: https://github.com/mcederle99/FairMSS.git).
In a recent study, Reinforcement Learning (RL) used in combination with many-objective search, has been shown to outperform alternative techniques (random search and many-objective search) for online testing of Deep Neural Network-enabled systems. The empirical evaluation of these techniques was conducted on a state-of-the-art Autonomous Driving System (ADS). This work is a replication and extension of that empirical study. Our replication shows that RL does not outperform pure random test generation in a comparison conducted under the same settings of the original study, but with no confounding factor coming from the way collisions are measured. Our extension aims at eliminating some of the possible reasons for the poor performance of RL observed in our replication: (1) the presence of reward components providing contrasting or useless feedback to the RL agent; (2) the usage of an RL algorithm (Q-learning) which requires discretization of an intrinsically continuous state space. Results show that our new RL agent is able to converge to an effective policy that outperforms random testing. Results also highlight other possible improvements, which open to further investigations on how to best leverage RL for online ADS testing.