In order for an autonomous robot to efficiently explore an unknown environment, it must account for uncertainty in sensor measurements, hazard assessment, localization, and motion execution. Making decisions for maximal reward in a stochastic setting requires value learning and policy construction over a belief space, i.e., probability distribution over all possible robot-world states. However, belief space planning in a large spatial environment over long temporal horizons suffers from severe computational challenges. Moreover, constructed policies must safely adapt to unexpected changes in the belief at runtime. This work proposes a scalable value learning framework, PLGRIM (Probabilistic Local and Global Reasoning on Information roadMaps), that bridges the gap between (i) local, risk-aware resiliency and (ii) global, reward-seeking mission objectives. Leveraging hierarchical belief space planners with information-rich graph structures, PLGRIM addresses large-scale exploration problems while providing locally near-optimal coverage plans. We validate our proposed framework with high-fidelity dynamic simulations in diverse environments and on physical robots in Martian-analog lava tubes.
相關內容
We consider the joint design and control of discrete-time stochastic dynamical systems over a finite time horizon. We formulate the problem as a multi-step optimization problem under uncertainty seeking to identify a system design and a control policy that jointly maximize the expected sum of rewards collected over the time horizon considered. The transition function, the reward function and the policy are all parametrized, assumed known and differentiable with respect to their parameters. We then introduce a deep reinforcement learning algorithm combining policy gradient methods with model-based optimization techniques to solve this problem. In essence, our algorithm iteratively approximates the gradient of the expected return via Monte-Carlo sampling and automatic differentiation and takes projected gradient ascent steps in the space of environment and policy parameters. This algorithm is referred to as Direct Environment and Policy Search (DEPS). We assess the performance of our algorithm in three environments concerned with the design and control of a mass-spring-damper system, a small-scale off-grid power system and a drone, respectively. In addition, our algorithm is benchmarked against a state-of-the-art deep reinforcement learning algorithm used to tackle joint design and control problems. We show that DEPS performs at least as well or better in all three environments, consistently yielding solutions with higher returns in fewer iterations. Finally, solutions produced by our algorithm are also compared with solutions produced by an algorithm that does not jointly optimize environment and policy parameters, highlighting the fact that higher returns can be achieved when joint optimization is performed.
Deep reinforcement learning provides a promising approach for text-based games in studying natural language communication between humans and artificial agents. However, the generalization still remains a big challenge as the agents depend critically on the complexity and variety of training tasks. In this paper, we address this problem by introducing a hierarchical framework built upon the knowledge graph-based RL agent. In the high level, a meta-policy is executed to decompose the whole game into a set of subtasks specified by textual goals, and select one of them based on the KG. Then a sub-policy in the low level is executed to conduct goal-conditioned reinforcement learning. We carry out experiments on games with various difficulty levels and show that the proposed method enjoys favorable generalizability.
The ability to autonomously navigate in unknown environments is important for mobile robots. The map is the core component to achieve this. Most map representations rely on drift-free state estimation and provide a global metric map to navigate. However, in large-scale real-world applications, it's hard to prohibit drifts and compose a globally consistent map quickly. In this paper, a novel representation named, HiTMap, is proposed to enhance the existing map representations. The central idea is to adopt a submap-based hierarchical topology rather than a global metric map so that only a local metric map is maintained for obstacle avoidance which ensures the lightweight of the representation. To guide the robots navigate into unknown spaces, frontiers are detected and attached to the map as an attribute. We also develop a path planning module to evaluate the feasibility and efficiency of our map representation. The system is validated in a simulation environment and a demonstration in the real world is conducted. In addition, the HiTMap is made available open-source.
The autonomous operation of small quadcopters moving at high speed in an unknown cluttered environment is a challenging task. Current works in the literature formulate it as a Sense-And-Avoid (SAA) problem and address it by either developing new sensing capabilities or small form-factor processors. However, the SAA, with the high-speed operation, remains an open problem. The significant complexity arises due to the computational latency, which is critical for fast-moving quadcopters. In this paper, a novel Fast Obstacle Avoidance Motion (FOAM) algorithm is proposed to perform SAA operations. FOAM is a low-latency perception-based algorithm that uses multi-sensor fusion of a monocular camera and a 2-D LIDAR. A 2-D probabilistic occupancy map of the sensing region is generated to estimate a free space for avoiding obstacles. Also, a local planner is used to navigate the high-speed quadcopter towards a given target location while avoiding obstacles. The performance evaluation of FOAM is evaluated in simulated environments in Gazebo and AIRSIM. Real-time implementation of the same has been presented in outdoor environments using a custom-designed quadcopter operating at a speed of $4.5$ m/s. The FOAM algorithm is implemented on a low-cost computing device to demonstrate its efficacy. The results indicate that FOAM enables a small quadcopter to operate at high speed in a cluttered environment efficiently.
Non-prehensile multi-object rearrangement is a robotic task of planning feasible paths and transferring multiple objects to their predefined target poses without grasping. It needs to consider how each object reaches the target and the order of object movement, which significantly deepens the complexity of the problem. To address these challenges, we propose a hierarchical policy to divide and conquer for non-prehensile multi-object rearrangement. In the high-level policy, guided by a designed policy network, the Monte Carlo Tree Search efficiently searches for the optimal rearrangement sequence among multiple objects, which benefits from imitation and reinforcement. In the low-level policy, the robot plans the paths according to the order of path primitives and manipulates the objects to approach the goal poses one by one. We verify through experiments that the proposed method can achieve a higher success rate, fewer steps, and shorter path length compared with the state-of-the-art.
The problem of multi-robot navigation of connectivity maintenance is challenging in multi-robot applications. This work investigates how to navigate a multi-robot team in unknown environments while maintaining connectivity. We propose a reinforcement learning (RL) approach to develop a decentralized policy, which is shared among multiple robots. Given range sensor measurements and the positions of other robots, the policy aims to generate control commands for navigation and preserve the global connectivity of the robot team. We incorporate connectivity concerns into the RL framework as constraints and introduce behavior cloning to reduce the exploration complexity of policy optimization. The policy is optimized with all transition data collected by multiple robots in random simulated scenarios. We validate the effectiveness of the proposed approach by comparing different combinations of connectivity constraints and behavior cloning. We also show that our policy can generalize to unseen scenarios in both simulation and holonomic robots experiments.
Our goal is to train control policies that generalize well to unseen environments. Inspired by the Distributionally Robust Optimization (DRO) framework, we propose DRAGEN - Distributionally Robust policy learning via Adversarial Generation of ENvironments - for iteratively improving robustness of policies to realistic distribution shifts by generating adversarial environments. The key idea is to learn a generative model for environments whose latent variables capture cost-predictive and realistic variations in environments. We perform DRO with respect to a Wasserstein ball around the empirical distribution of environments by generating realistic adversarial environments via gradient ascent on the latent space. We demonstrate strong Out-of-Distribution (OoD) generalization in simulation for (i) swinging up a pendulum with onboard vision and (ii) grasping realistic 2D/3D objects. Grasping experiments on hardware demonstrate better sim2real performance compared to domain randomization.
We study the offline meta-reinforcement learning (OMRL) problem, a paradigm which enables reinforcement learning (RL) algorithms to quickly adapt to unseen tasks without any interactions with the environments, making RL truly practical in many real-world applications. This problem is still not fully understood, for which two major challenges need to be addressed. First, offline RL usually suffers from bootstrapping errors of out-of-distribution state-actions which leads to divergence of value functions. Second, meta-RL requires efficient and robust task inference learned jointly with control policy. In this work, we enforce behavior regularization on learned policy as a general approach to offline RL, combined with a deterministic context encoder for efficient task inference. We propose a novel negative-power distance metric on bounded context embedding space, whose gradients propagation is detached from the Bellman backup. We provide analysis and insight showing that some simple design choices can yield substantial improvements over recent approaches involving meta-RL and distance metric learning. To the best of our knowledge, our method is the first model-free and end-to-end OMRL algorithm, which is computationally efficient and demonstrated to outperform prior algorithms on several meta-RL benchmarks.
Recently, deep multiagent reinforcement learning (MARL) has become a highly active research area as many real-world problems can be inherently viewed as multiagent systems. A particularly interesting and widely applicable class of problems is the partially observable cooperative multiagent setting, in which a team of agents learns to coordinate their behaviors conditioning on their private observations and commonly shared global reward signals. One natural solution is to resort to the centralized training and decentralized execution paradigm. During centralized training, one key challenge is the multiagent credit assignment: how to allocate the global rewards for individual agent policies for better coordination towards maximizing system-level's benefits. In this paper, we propose a new method called Q-value Path Decomposition (QPD) to decompose the system's global Q-values into individual agents' Q-values. Unlike previous works which restrict the representation relation of the individual Q-values and the global one, we leverage the integrated gradient attribution technique into deep MARL to directly decompose global Q-values along trajectory paths to assign credits for agents. We evaluate QPD on the challenging StarCraft II micromanagement tasks and show that QPD achieves the state-of-the-art performance in both homogeneous and heterogeneous multiagent scenarios compared with existing cooperative MARL algorithms.
Despite deep reinforcement learning has recently achieved great successes, however in multiagent environments, a number of challenges still remain. Multiagent reinforcement learning (MARL) is commonly considered to suffer from the problem of non-stationary environments and exponentially increasing policy space. It would be even more challenging to learn effective policies in circumstances where the rewards are sparse and delayed over long trajectories. In this paper, we study Hierarchical Deep Multiagent Reinforcement Learning (hierarchical deep MARL) in cooperative multiagent problems with sparse and delayed rewards, where efficient multiagent learning methods are desperately needed. We decompose the original MARL problem into hierarchies and investigate how effective policies can be learned hierarchically in synchronous/asynchronous hierarchical MARL frameworks. Several hierarchical deep MARL architectures, i.e., Ind-hDQN, hCom and hQmix, are introduced for different learning paradigms. Moreover, to alleviate the issues of sparse experiences in high-level learning and non-stationarity in multiagent settings, we propose a new experience replay mechanism, named as Augmented Concurrent Experience Replay (ACER). We empirically demonstrate the effects and efficiency of our approaches in several classic Multiagent Trash Collection tasks, as well as in an extremely challenging team sports game, i.e., Fever Basketball Defense.
小貼士
登錄享
相關主題
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191