DCOB: Action space for reinforcement learning of high DoF robots

Reinforcement learning (RL) for robot control is an important technology for future robots since it enables us to design a robot’s behavior using the reward function. However, RL for high degree-of-freedom robot control is still an open issue. This paper proposes a discrete action space DCOB which is generated from the basis functions (BFs) given to approximate a value function. The remarkable feature is that, by reducing the number of BFs to enable the robot to learn quickly the value function, the size of DCOB is also reduced, which improves the learning speed. In addition, a method WF-DCOB is proposed to enhance the performance, where wire-fitting is utilized to search for continuous actions around each discrete action of DCOB. We apply the proposed methods to motion learning tasks of a simulated humanoid robot and a real spider robot. The experimental results demonstrate outstanding performance.     

Adaptive control of a looper-like robot based on the CPG-actor-critic method

Adaptability to the environment is crucial for mobile robots, because the circumstances, including the body of the robot, may change. A robot with a large number of degrees of freedom possesses the potential to adapt to such circumstances, but it is difficult to design a good controller for such a robot. We previously proposed a reinforcement learning (RL) method called the CPG actor-critic method, and applied it to the automatic acquisition of vermicular locomotion of a looper-like robot through computer simulations. In this study, we developed a looper-like robot and applied our RL method to the control of this robot. Experimental results demonstrate fast acquisition of a vermicular forward motion, supporting the real applicability of our method. This work was presented in part at the 12th International Symposium on Artificial Life and Robotics, Oita, Japan, January 25–27, 2007     

Dyna-Q-based vector direction for path planning problem of autonomous mobile robots in unknown environments

Reinforcement learning (RL) is a popular method for solving the path planning problem of autonomous mobile robots in unknown environments. However, the primary difficulty faced by learning robots using the RL method is that they learn too slowly in obstacle-dense environments. To more efficiently solve the path planning problem of autonomous mobile robots in such environments, this paper presents a novel approach in which the robot’s learning process is divided into two phases. The first one is to accelerate the learning process for obtaining an optimal policy by developing the well-known Dyna-Q algorithm that trains the robot in learning actions for avoiding obstacles when following the vector direction. In this phase, the robot’s position is represented as a uniform grid. At each time step, the robot performs an action to move to one of its eight adjacent cells, so the path obtained from the optimal policy may be longer than the true shortest path. The second one is to train the robot in learning a collision-free smooth path for decreasing the number of the heading changes of the robot. The simulation results show that the proposed approach is efficient for the path planning problem of autonomous mobile robots in unknown environments with dense obstacles.     

TEXPLORE: real-time sample-efficient reinforcement learning for robots

The use of robots in society could be expanded by using reinforcement learning (RL) to allow robots to learn and adapt to new situations online. RL is a paradigm for learning sequential decision making tasks, usually formulated as a Markov Decision Process (MDP). For an RL algorithm to be practical for robotic control tasks, it must learn in very few samples, while continually taking actions in real-time. In addition, the algorithm must learn efficiently in the face of noise, sensor/actuator delays, and continuous state features. In this article, we present texplore, the first algorithm to address all of these challenges together. texplore is a model-based RL method that learns a random forest model of the domain which generalizes dynamics to unseen states. The agent explores states that are promising for the final policy, while ignoring states that do not appear promising. With sample-based planning and a novel parallel architecture, texplore can select actions continually in real-time whenever necessary. We empirically evaluate the importance of each component of texplore in isolation and then demonstrate the complete algorithm learning to control the velocity of an autonomous vehicle in real-time.     

Heuristically Accelerated Reinforcement Learning by Means of Case-Based Reasoning and Transfer Learning

Reinforcement Learning (RL) is a well-known technique for learning the solutions of control problems from the interactions of an agent in its domain. However, RL is known to be inefficient in problems of the real-world where the state space and the set of actions grow up fast. Recently, heuristics, case-based reasoning (CBR) and transfer learning have been used as tools to accelerate the RL process. This paper investigates a class of algorithms called Transfer Learning Heuristically Accelerated Reinforcement Learning (TLHARL) that uses CBR as heuristics within a transfer learning setting to accelerate RL. The main contributions of this work are the proposal of a new TLHARL algorithm based on the traditional RL algorithm Q(λ) and the application of TLHARL on two distinct real-robot domains: a robot soccer with small-scale robots and the humanoid-robot stability learning. Experimental results show that our proposed method led to a significant improvement of the learning rate in both domains.     

Interactive Teaching for Vision-Based Mobile Robots: A Sensory-Motor Approach

For the last decade, we have been developing a vision-based architecture for mobile robot navigation. Using our bio-inspired model of navigation, robots can perform sensory-motor tasks in real time in unknown indoor as well as outdoor environments. We address here the problem of autonomous incremental learning of a sensory-motor task, demonstrated by an operator guiding a robot. The proposed system allows for semisupervision of task learning and is able to adapt the environmental partitioning to the complexity of the desired behavior. A real dialogue based on actions emerges from the interactive teaching. The interaction leads the robot to autonomously build a precise sensory-motor dynamics that approximates the behavior of the teacher. The usability of the system is highlighted by experiments on real robots, in both indoor and outdoor environments. Accuracy measures are also proposed in order to evaluate the learned behavior as compared to the expected behavioral attractor. These measures, used first in a real experiment and then in a simulated experiment, demonstrate how a real interaction between the teacher and the robot influences the learning process.     

基于动作选择级的多机器人协作

在多机器人环境中,由于每个机器人动作选择的重叠现象,让机器人之间的协作变得很差.提出了一个方法用于确定动作选择级别.在此基础上,可以很好地控制多机器人的协作行为的获取.首先,定义了用于动作选择级优先级的8个级别,这8个级别相应的映射到8个动作子空间.然后,利用局部势场法,每个机器人的动作选择优先级被计算出来,并且因此,每个机器人获得了各自需要搜索的动作子空间.在动作子空间中,每个机器人利用加强学习方法来选择一个适当的动作.最终,把该方法用于机器人足球比赛的机器人局部协作训练中.试验的效果在仿真和实际比赛中得到了证实.     

MOSAIC for multiple-reward environments

Reinforcement learning (RL) can provide a basic framework for autonomous robots to learn to control and maximize future cumulative rewards in complex environments. To achieve high performance, RL controllers must consider the complex external dynamics for movements and task (reward function) and optimize control commands. For example, a robot playing tennis and squash needs to cope with the different dynamics of a tennis or squash racket and such dynamic environmental factors as the wind. In addition, this robot has to tailor its tactics simultaneously under the rules of either game. This double complexity of the external dynamics and reward function sometimes becomes more complex when both the multiple dynamics and multiple reward functions switch implicitly, as in the situation of a real (multi-agent) game of tennis where one player cannot observe the intention of her opponents or her partner. The robot must consider its opponent's and its partner's unobservable behavioral goals (reward function). In this article, we address how an RL agent should be designed to handle such double complexity of dynamics and reward. We have previously proposed modular selection and identification for control (MOSAIC) to cope with nonstationary dynamics where appropriate controllers are selected and learned among many candidates based on the error of its paired dynamics predictor: the forward model. Here we extend this framework for RL and propose MOSAIC-MR architecture. It resembles MOSAIC in spirit and selects and learns an appropriate RL controller based on the RL controller's TD error using the errors of the dynamics (the forward model) and the reward predictors. Furthermore, unlike other MOSAIC variants for RL, RL controllers are not a priori paired with the fixed predictors of dynamics and rewards. The simulation results demonstrate that MOSAIC-MR outperforms other counterparts because of this flexible association ability among RL controllers, forward models, and reward predictors.     

自主机器人的强化学习研究进展

虽然基于行为控制的自主机器人具有较高的鲁棒性,但其对于动态环境缺乏必要的自 适应能力．强化学习方法使机器人可以通过学习来完成任务,而无需设计者完全预先规定机 器人的所有动作,它是将动态规划和监督学习结合的基础上发展起来的一种新颖的学习方法 ,它通过机器人与环境的试错交互,利用来自成功和失败经验的奖励和惩罚信号不断改进机 器人的性能,从而达到目标,并容许滞后评价．由于其解决复杂问题的突出能力,强化学习 已成为一种非常有前途的机器人学习方法．本文系统论述了强化学习方法在自主机器人中的 研究现状,指出了存在的问题,分析了几种问题解决途径,展望了未来发展趋势．     

SSPQL: Stochastic shortest path-based Q-learning

Reinforcement learning (RL) has been widely used as a mechanism for autonomous robots to learn state-action pairs by interacting with their environment. However, most RL methods usually suffer from slow convergence when deriving an optimum policy in practical applications. To solve this problem, a stochastic shortest path-based Q-learning (SSPQL) is proposed, combining a stochastic shortest path-finding method with Q-learning, a well-known model-free RL method. The rationale is, if a robot has an internal state-transition model which is incrementally learnt, then the robot can infer the local optimum policy by using a stochastic shortest path-finding method. By increasing state-action pair values comprising of these local optimum policies, a robot can then reach a goal quickly and as a result, this process can enhance convergence speed. To demonstrate the validity of this proposed learning approach, several experimental results are presented in this paper.     

A morphable template framework for robot learning by demonstration: Integrating one-shot and incremental learning approaches

Robot learning by demonstration is key to bringing robots into daily social environments to interact with and learn from human and other agents. However, teaching a robot to acquire new knowledge is a tedious and repetitive process and often restrictive to a specific setup of the environment. We propose a template-based learning framework for robot learning by demonstration to address both generalisation and adaptability. This novel framework is based upon a one-shot learning model integrated with spectral clustering and an online learning model to learn and adapt actions in similar scenarios. A set of statistical experiments is used to benchmark the framework components and shows that this approach requires no extensive training for generalisation and can adapt to environmental changes flexibly. Two real-world applications of an iCub humanoid robot playing the tic-tac-toe game and soldering a circuit board are used to demonstrate the relative merits of the framework.     

Robots learn to dance through interaction with humans

In this paper, we investigated an approach for robots to learn to adapt dance actions to human’s preferences through interaction and feedback. Human’s preferences were extracted by analysing the common action patterns with positive or negative feedback from the human during robot dancing. By using a buffering technique to store the dance actions before a feedback, each individual’s preferences can be extracted even when a reward is received late. The extracted preferred dance actions from different people were then combined to generate improved dance sequences, i.e. performing more of what was preferred and less of that was not preferred. Together with Softmax action-selection method, the Sarsa reinforcement learning algorithm was used as the underlining learning algorithm and to effectively control the trade-off between exploitation of the learnt dance skills and exploration of new dance actions. The results showed that the robot learnt, using interactive reinforcement learning, the preferences of human partners, and the dance improved with the extracted preferences from more human partners.     

Real-world reinforcement learning for autonomous humanoid robot docking

Reinforcement learning (RL) is a biologically supported learning paradigm, which allows an agent to learn through experience acquired by interaction with its environment. Its potential to learn complex action sequences has been proven for a variety of problems, such as navigation tasks. However, the interactive randomized exploration of the state space, common in reinforcement learning, makes it difficult to be used in real-world scenarios. In this work we describe a novel real-world reinforcement learning method. It uses a supervised reinforcement learning approach combined with Gaussian distributed state activation. We successfully tested this method in two real scenarios of humanoid robot navigation: first, backward movements for docking at a charging station and second, forward movements to prepare grasping. Our approach reduces the required learning steps by more than an order of magnitude, and it is robust and easy to be integrated into conventional RL techniques.     

Learning proactive behavior for interactive social robots

Learning human–robot interaction logic from example interaction data has the potential to leverage “big data” to reduce the effort and time spent on designing interaction logic or crafting interaction content. Previous work has demonstrated techniques by which a robot can learn motion and speech behaviors from non-annotated human–human interaction data, but these techniques only enable a robot to respond to human-initiated inputs, and do not enable the robot to proactively initiate interaction. In this work, we propose a method for learning both human-initiated and robot-initiated behavior for a social robot from human–human example interactions, which we demonstrate for a shopkeeper interacting with a customer in a camera shop scenario. This was achieved by extending an existing technique by (1) introducing a concept of a customer yield action, (2) incorporating interaction history, represented by sequences of discretized actions, as inputs for training and generating robot behavior, and (3) using an “attention mechanism” in our learning system for training robot behaviors, that learns which parts of the interaction history are more important for generating robot behaviors. The proposed method trains a robot to generate multimodal actions, consisting of speech and locomotion behaviors. We compared this study with the previous technique in two ways. Cross-validation on the training data showed higher social appropriateness of predicted behaviors using the proposed technique, and a user study of live interaction with a robot showed that participants perceived the proposed technique to produce behaviors that were more proactive, socially-appropriate, and better in overall quality.     

A distributed and morphology-independent strategy for adaptive locomotion in self-reconfigurable modular robots

In this paper, we present a distributed reinforcement learning strategy for morphology-independent life-long gait learning for modular robots. All modules run identical controllers that locally and independently optimize their action selection based on the robot’s velocity as a global, shared reward signal. We evaluate the strategy experimentally mainly on simulated, but also on physical, modular robots. We find that the strategy: (i) for six of seven configurations (3–12 modules) converge in 96% of the trials to the best known action-based gaits within 15 min, on average, (ii) can be transferred to physical robots with a comparable performance, (iii) can be applied to learn simple gait control tables for both M-TRAN and ATRON robots, (iv) enables an 8-module robot to adapt to faults and changes in its morphology, and (v) can learn gaits for up to 60 module robots but a divergence effect becomes substantial from 20–30 modules. These experiments demonstrate the advantages of a distributed learning strategy for modular robots, such as simplicity in implementation, low resource requirements, morphology independence, reconfigurability, and fault tolerance.     

Improving human robot collaboration through Force/Torque based learning for object manipulation

Human–Robot Collaboration (HRC) is a term used to describe tasks in which robots and humans work together to achieve a goal. Unlike traditional industrial robots, collaborative robots need to be adaptive; able to alter their approach to better suit the situation and the needs of the human partner. As traditional programming techniques can struggle with the complexity required, an emerging approach is to learn a skill by observing human demonstration and imitating the motions; commonly known as Learning from Demonstration (LfD). In this work, we present a LfD methodology that combines an ensemble machine learning algorithm (i.e. Random Forest (RF)) with stochastic regression, using haptic information captured from human demonstration. The capabilities of the proposed method are evaluated using two collaborative tasks; co-manipulation of an object (where the human provides the guidance but the robot handles the objects weight) and collaborative assembly of simple interlocking parts. The proposed method is shown to be capable of imitation learning; interpreting human actions and producing equivalent robot motion across a diverse range of initial and final conditions. After verifying that ensemble machine learning can be utilised for real robotics problems, we propose a further extension utilising Weighted Random Forest (WRF) that attaches weights to each tree based on its performance. It is then shown that the WRF approach outperforms RF in HRC tasks.     

A reinforcement learning with evolutionary state recruitment strategy for autonomous mobile robots control

In recent robotics fields, much attention has been focused on utilizing reinforcement learning (RL) for designing robot controllers, since environments where the robots will be situated in should be unpredictable for human designers in advance. However there exist some difficulties. One of them is well known as ‘curse of dimensionality problem’. Thus, in order to adopt RL for complicated systems, not only ‘adaptability’ but also ‘computational efficiencies’ should be taken into account. The paper proposes an adaptive state recruitment strategy for NGnet-based actor-critic RL. The strategy enables the learning system to rearrange/divide its state space gradually according to the task complexity and the progress of learning. Some simulation results and real robot implementations show the validity of the method.     

Pneumatic artificial muscle-driven robot control using local update reinforcement learning

In this study, a new value function based Reinforcement learning (RL) algorithm, Local Update Dynamic Policy Programming (LUDPP), is proposed. It exploits the nature of smooth policy update using Kullback–Leibler divergence to update its value function locally and considerably reduces the computational complexity. We firstly investigated the learning performance of LUDPP and other algorithms without smooth policy update for tasks of pendulum swing up and n DOFs manipulator reaching in simulation. Only LUDPP could efficiently and stably learn good control policies in high dimensional systems with limited number of training samples. In real word application, we applied LUDPP to control Pneumatic Artificial Muscles (PAMs) driven robots without the knowledge of model which is challenging for traditional methods due to the high nonlinearities of PAM’s air pressure dynamics and mechanical structure. LUDPP successfully achieved one finger control of Shadow Dexterous Hand, a PAM-driven humanoid robot hand, with far lower computational resource compared with other conventional value function based RL algorithms.     

Evolutionary Development of Hierarchical Learning Structures

Hierarchical reinforcement learning (RL) algorithms can learn a policy faster than standard RL algorithms. However, the applicability of hierarchical RL algorithms is limited by the fact that the task decomposition has to be performed in advance by the human designer. We propose a Lamarckian evolutionary approach for automatic development of the learning structure in hierarchical RL. The proposed method combines the MAXQ hierarchical RL method and genetic programming (GP). In the MAXQ framework, a subtask can optimize the policy independently of its parent task's policy, which makes it possible to reuse learned policies of the subtasks. In the proposed method, the MAXQ method learns the policy based on the task hierarchies obtained by GP, while the GP explores the appropriate hierarchies using the result of the MAXQ method. To show the validity of the proposed method, we have performed simulation experiments for a foraging task in three different environmental settings. The results show strong interconnection between the obtained learning structures and the given task environments. The main conclusion of the experiments is that the GP can find a minimal strategy, i.e., a hierarchy that minimizes the number of primitive subtasks that can be executed for each type of situation. The experimental results for the most challenging environment also show that the policies of the subtasks can continue to improve, even after the structure of the hierarchy has been evolutionary stabilized, as an effect of Lamarckian mechanisms     

Experience Learning From Basic Patterns for Efficient Robot Navigation in Indoor Environments

In this paper we propose a machine learning technique for real-time robot path planning for an autonomous robot in a planar environment with obstacles where the robot possess no a priori map of its environment. Our main insight in this paper is that a robot’s path planning times can be significantly reduced if it can refer to previous maneuvers it used to avoid obstacles during earlier missions, and adapt that information to avoid obstacles during its current navigation. We propose an online path planning algorithm called LearnerRRT that utilizes a pattern matching technique called Sample Consensus Initial Alignment (SAC-IA) in combination with an experience-based learning technique to adapt obstacle boundary patterns encountered in previous environments to the current scenario followed by corresponding adaptations in the obstacle-avoidance paths. Our proposed algorithm LearnerRRT works as a learning-based reactive path planning technique which enables robots to improve their overall path planning performance by locally improving maneuvers around commonly encountered obstacle patterns by accessing previously accumulated environmental information. We have conducted several experiments in simulations and hardware to verify the performance of the LearnerRRT algorithm and compared it with a state-of-the-art sampling-based planner. LearnerRRT on average takes approximately 10% of the planning time and 14% of the total time taken by the sampling-based planner to solve the same navigation task based on simulation results and takes only 33% of the planning time, 46% of total time and 95% of total distance compared to the sampling-based planner based on our hardware results.