Category Archives: Applications Of Reinforcement Learning To Robots

Using proprioceptive, internal perceptions, in robots, with RL

Agnese Augello, Salvatore Gaglio, Ignazio Infantino, Umberto Maniscalco, Giovanni Pilato, Filippo Vella, Roboception and adaptation in a cognitive robot, Robotics and Autonomous Systems, Volume 164, 2023 DOI: 10.1016/j.robot.2023.104400.

In robotics, perception is usually oriented at understanding what is happening in the external world, while few works pay attention to what is occurring in the robot\u2019s body. In this work, we propose an artificial somatosensory system, embedded in a cognitive architecture, that enables a robot to perceive the sensations from its embodiment while executing a task. We called these perceptions roboceptions, and they let the robot act according to its own physical needs in addition to the task demands. Physical information is processed by the robot to behave in a balanced way, determining the most appropriate trade-off between the achievement of the task and its well being. The experiments show the integration of information from the somatosensory system and the choices that lead to the accomplishment of the task.

Q-learning with a variation of e-greedy to learn the optimal management of energy in autonomous vehicles navigation

Mojgan Fayyazi, Monireh Abdoos, Duong Phan, Mohsen Golafrouz, Mahdi Jalili, Reza N. Jazar, Reza Langari, Hamid Khayyam, Real-time self-adaptive Q-learning controller for energy management of conventional autonomous vehicles, Expert Systems with Applications, Volume 222, 2023 DOI: 10.1016/j.eswa.2023.119770.

Reducing emissions and energy consumption of autonomous vehicles is critical in the modern era. This paper presents an intelligent energy management system based on Reinforcement Learning (RL) for conventional autonomous vehicles. Furthermore, in order to improve the efficiency, a new exploration strategy is proposed to replace the traditional decayed \u03b5-greedy strategy in the Q-learning algorithm associated with RL. Unlike traditional Q-learning algorithms, the proposed self-adaptive Q-learning (SAQ-learning) can be applied in real-time. The learning capability of the controllers can help the vehicle deal with unknown situations in real-time. Numerical simulations show that compared to other controllers, Q-learning and SAQ-learning controllers can generate the desired engine torque based on the vehicle road power demand and control the air/fuel ratio by changing the throttle angle efficiently in real-time. Also, the proposed real-time SAQ-learning is shown to improve the operational time by 23% compared to standard Q-learning. Our simulations reveal the effectiveness of the proposed control system compared to other methods, namely dynamic programming and fuzzy logic methods.

Review of RL applied to robotic manipulation

��igo Elguea-Aguinaco, Antonio Serrano-Mu�oz, Dimitrios Chrysostomou, Ibai Inziarte-Hidalgo, Simon B�gh, Nestor Arana-Arexolaleiba, A review on reinforcement learning for contact-rich robotic manipulation tasks, Robotics and Computer-Integrated Manufacturing, Volume 81, 2023 DOI: 10.1016/j.rcim.2022.102517.

Research and application of reinforcement learning in robotics for contact-rich manipulation tasks have exploded in recent years. Its ability to cope with unstructured environments and accomplish hard-to-engineer behaviors has led reinforcement learning agents to be increasingly applied in real-life scenarios. However, there is still a long way ahead for reinforcement learning to become a core element in industrial applications. This paper examines the landscape of reinforcement learning and reviews advances in its application in contact-rich tasks from 2017 to the present. The analysis investigates the main research for the most commonly selected tasks for testing reinforcement learning algorithms in both rigid and deformable object manipulation. Additionally, the trends around reinforcement learning associated with serial manipulators are explored as well as the various technological challenges that this machine learning control technique currently presents. Lastly, based on the state-of-the-art and the commonalities among the studies, a framework relating the main concepts of reinforcement learning in contact-rich manipulation tasks is proposed. The final goal of this review is to support the robotics community in future development of systems commanded by reinforcement learning, discuss the main challenges of this technology and suggest future research directions in the domain.

Including safety learning in RL for improving the sim-to-lab gap

Kai-Chieh Hsu, Allen Z. Ren, Duy P. Nguyen, Anirudha Majumdar, Jaime F. Fisac, Sim-to-Lab-to-Real: Safe reinforcement learning with shielding and generalization guarantees, Artificial Intelligence, Volume 314, 2023 DOI: 10.1016/j.artint.2022.103811.

Safety is a critical component of autonomous systems and remains a challenge for learning-based policies to be utilized in the real world. In particular, policies learned using reinforcement learning often fail to generalize to novel environments due to unsafe behavior. In this paper, we propose Sim-to-Lab-to-Real to bridge the reality gap with a probabilistically guaranteed safety-aware policy distribution. To improve safety, we apply a dual policy setup where a performance policy is trained using the cumulative task reward and a backup (safety) policy is trained by solving the Safety Bellman Equation based on Hamilton-Jacobi (HJ) reachability analysis. In Sim-to-Lab transfer, we apply a supervisory control scheme to shield unsafe actions during exploration; in Lab-to-Real transfer, we leverage the Probably Approximately Correct (PAC)-Bayes framework to provide lower bounds on the expected performance and safety of policies in unseen environments. Additionally, inheriting from the HJ reachability analysis, the bound accounts for the expectation over the worst-case safety in each environment. We empirically study the proposed framework for ego-vision navigation in two types of indoor environments with varying degrees of photorealism. We also demonstrate strong generalization performance through hardware experiments in real indoor spaces with a quadrupedal robot. See for supplementary material.

Adaptation of industrial robots to variations in tasks through RL

Tian Yu, Qing Chang, User-guided motion planning with reinforcement learning for human-robot collaboration in smart manufacturing, Expert Systems with Applications, Volume 209, 2022 DOI: 10.1016/j.eswa.2022.118291.

In today\u2019s manufacturing system, robots are expected to perform increasingly complex manipulation tasks in collaboration with humans. However, current industrial robots are still largely preprogrammed with very little autonomy and still required to be reprogramed by robotics experts for even slightly changed tasks. Therefore, it is highly desirable that robots can adapt to certain task changes with motion planning strategies to easily work with non-robotic experts in manufacturing environments. In this paper, we propose a user-guided motion planning algorithm in combination with reinforcement learning (RL) method to enable robots automatically generate their motion plans for new tasks by learning from a few kinesthetic human demonstrations. Features of common human demonstrated tasks in a specific application environment, e.g., desk assembly or warehouse loading/unloading are abstracted and saved in a library. The definition of semantical similarity between features in the library and features of a new task is proposed and further used to construct the reward function in RL. To achieve an adaptive motion plan facing task changes or new task requirements, features embedded in the library are mapped to appropriate task segments based on the trained motion planning policy using Q-learning. A new task can be either learned as a combination of a few features in the library or a requirement for further human demonstration if the current library is insufficient for the new task. We evaluate our approach on a 6 DOF UR5e robot on multiple tasks and scenarios and show the effectiveness of our method with respect to different scenarios.

On the extended use of RL for navigation in UAVs

Fadi AlMahamid, Katarina Grolinger, Autonomous Unmanned Aerial Vehicle navigation using Reinforcement Learning: A systematic review, Engineering Applications of Artificial Intelligence, Volume 115, 2022 DOI: 10.1016/j.engappai.2022.105321.

There is an increasing demand for using Unmanned Aerial Vehicle (UAV), known as drones, in different applications such as packages delivery, traffic monitoring, search and rescue operations, and military combat engagements. In all of these applications, the UAV is used to navigate the environment autonomously \u2014 without human interaction, perform specific tasks and avoid obstacles. Autonomous UAV navigation is commonly accomplished using Reinforcement Learning (RL), where agents act as experts in a domain to navigate the environment while avoiding obstacles. Understanding the navigation environment and algorithmic limitations plays an essential role in choosing the appropriate RL algorithm to solve the navigation problem effectively. Consequently, this study first identifies the main UAV navigation tasks and discusses navigation frameworks and simulation software. Next, RL algorithms are classified and discussed based on the environment, algorithm characteristics, abilities, and applications in different UAV navigation problems, which will help the practitioners and researchers select the appropriate RL algorithms for their UAV navigation use cases. Moreover, identified gaps and opportunities will drive UAV navigation research.

Hierarchical RL with diverse methods integrated in the framework

Ye Zhou, Hann Woei Ho, Online robot guidance and navigation in non-stationary environment with hybrid Hierarchical Reinforcement Learning, Engineering Applications of Artificial Intelligence, Volume 114, 2022 DOI: 10.1016/j.engappai.2022.105152.

Hierarchical Reinforcement Learning (HRL) provides an option to solve complex guidance and navigation problems with high-dimensional spaces, multiple objectives, and a large number of states and actions. The current HRL methods often use the same or similar reinforcement learning methods within one application so that multiple objectives can be easily combined. Since there is not a single learning method that can benefit all targets, hybrid Hierarchical Reinforcement Learning (hHRL) was proposed to use various methods to optimize the learning with different types of information and objectives in one application. The previous hHRL method, however, requires manual task-specific designs, which involves engineers\u2019 preferences and may impede its transfer learning ability. This paper, therefore, proposes a systematic online guidance and navigation method under the framework of hHRL, which generalizes training samples with a function approximator, decomposes the state space automatically, and thus does not require task-specific designs. The simulation results indicate that the proposed method is superior to the previous hHRL method, which requires manual decomposition, in terms of the convergence rate and the learnt policy. It is also shown that this method is generally applicable to non-stationary environments changing over episodes and over time without the loss of efficiency even with noisy state information.

Hybridizing model-free and model-based in continuous RL, and a nice review of current research and benchmarks in robotics

Pinosky A, Abraham I, Broad A, Argall B, Murphey TD. Hybrid control for combining model-based and model-free reinforcement learning The International Journal of Robotics Research. 2023;42(6):337-355 DOI: 10.1177/02783649221083331.

We develop an approach to improve the learning capabilities of robotic systems by combining learned predictive models with experience-based state-action policy mappings. Predictive models provide an understanding of the task and the dynamics, while experience-based (model-free) policy mappings encode favorable actions that override planned actions. We refer to our approach of systematically combining model-based and model-free learning methods as hybrid learning. Our approach efficiently learns motor skills and improves the performance of predictive models and experience-based policies. Moreover, our approach enables policies (both model-based and model-free) to be updated using any off-policy reinforcement learning method. We derive a deterministic method of hybrid learning by optimally switching between learning modalities. We adapt our method to a stochastic variation that relaxes some of the key assumptions in the original derivation. Our deterministic and stochastic variations are tested on a variety of robot control benchmark tasks in simulation as well as a hardware manipulation task. We extend our approach for use with imitation learning methods, where experience is provided through demonstrations, and we test the expanded capability with a real-world pick-and-place task. The results show that our method is capable of improving the performance and sample efficiency of learning motor skills in a variety of experimental domains.

POMDPs in robotics: QMDP-Net as a counterpart for the Partially Observable Markov Decision Process (POMDP) whose transition, observation, and reward functions are initially unknown

Collins N, Kurniawati H. Locally connected interrelated network: A forward propagation primitive, The International Journal of Robotics Research. 2023;42(6):371-384 DOI: 10.1177/02783649221093092.

End-to-end learning for planning is a promising approach for finding good robot strategies in situations where the state transition, observation, and reward functions are initially unknown. Many neural network architectures for this approach have shown positive results. Across these networks, seemingly small components have been used repeatedly in different architectures, which means improving the efficiency of these components has great potential to improve the overall performance of the network. This paper aims to improve one such component: The forward propagation module. In particular, we propose Locally Connected Interrelated Network (LCI-Net) \u2013 a novel type of locally connected layer with unshared but interrelated weights \u2013 to improve the efficiency of learning stochastic transition models for planning and propagating information via the learned transition models. LCI-Net is a small differentiable neural network module that can be plugged into various existing architectures. For evaluation purposes, we apply LCI-Net to VIN and QMDP-Net. VIN is an end-to-end neural network for solving Markov Decision Processes (MDPs) whose transition and reward functions are initially unknown, while QMDP-Net is its counterpart for the Partially Observable Markov Decision Process (POMDP) whose transition, observation, and reward functions are initially unknown. Simulation tests on benchmark problems involving 2D and 3D navigation and grasping indicate promising results: Changing only the forward propagation module alone with LCI-Net improves VIN\u2019s and QMDP-Net generalisation capability by more than 3� and 10�, respectively.

Modifications of Q-learning for better learning of robot navigation

Ee Soong Low, Pauline Ong, Cheng Yee Low, Rosli Omar, Modified Q-learning with distance metric and virtual target on path planning of mobile robot, Expert Systems with Applications, Volume 199, 2022, DOI: 10.1016/j.eswa.2022.117191.

Path planning is an essential element in mobile robot navigation. One of the popular path planners is Q-learning \u2013 a type of reinforcement learning that learns with little or no prior knowledge of the environment. Despite the successful implementation of Q-learning reported in numerous studies, its slow convergence associated with the curse of dimensionality may limit the performance in practice. To solve this problem, an Improved Q-learning (IQL) with three modifications is introduced in this study. First, a distance metric is added to Q-learning to guide the agent moves towards the target. Second, the Q function of Q-learning is modified to overcome dead-ends more effectively. Lastly, the virtual target concept is introduced in Q-learning to bypass dead-ends. Experimental results across twenty types of navigation maps show that the proposed strategies accelerate the learning speed of IQL in comparison with the Q-learning. Besides, performance comparison with seven well-known path planners indicates its efficiency in terms of the path smoothness, time taken, shortest distance and total distance used.