Robot navigation in a crowd by integrating deep reinforcement learning and online planning

نویسندگان

چکیده

Navigating mobile robots along time-efficient and collision-free paths in crowds is still an open challenging problem. The key to build a profound understanding of the crowd for robots, which basis proactive foresighted policy. However, since interaction mechanisms among pedestrians are complex sophisticated, it difficult describe model them accurately. For excellent approximation capability deep neural networks, reinforcement learning promising solution current model-free learning-based approaches navigation always neglect planning lead reactive collision avoidance policies shortsighted behaviors. Meanwhile, most model-based based on state values, imposing substantial computational burden. To address these problems, we propose graph-based method, social double dueling Q-network (SG-D3QN), that (i) introduces attention mechanism extract efficient graph representation crowd-robot state, (ii) extends previous value approximator state-action approximator, (iii) further optimizes with simulated experiences generated by learned environment model, (iv) then proposes human-like decision-making process integrating online planning. Experimental results indicate our approach helps robot understand achieves high success rate more than 0.99 task. Compared state-of-the-art algorithms, better performance. Furthermore, process, incurs less half cost.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

application of reinforcement learning for navigation of a planar snake robot in serpentine locomotion

this article presents an implementation of a reinforcement learning (rl) method for a snake like robot navigation. the paper starts with developing kinematics and dynamics model of a snake robot in serpentine locomotion followed by performing simulation and finishes with actual experimentation. first, gibbs-appell's method is used to obtain the robot dynamics. the robot is also modeled in simme...

متن کامل

Hierarchical Reinforcement Learning for Robot Navigation

For complex tasks, such as manipulation and robot navigation, reinforcement learning (RL) is well-known to be difficult due to the curse of dimensionality. To overcome this complexity and making RL feasible, hierarchical RL (HRL) has been suggested. The basic idea of HRL is to divide the original task into elementary subtasks, which can be learned using RL. In this paper, we propose a HRL archi...

متن کامل

Vision-based Reinforcement Learning for Robot Navigation

In this paper, we present a novel vision-based learning approach for autonomous robot navigation. A hybrid statemapping model, which combines the merits of both static and dynamic state assigning strategies, is proposed to solve the problem of state organization in navigation-learning tasks. Specifically, the continuous feature space, which could be very large in general, is firstly mapped to a...

متن کامل

Reinforcement learning-based mobile robot navigation

In recent decades, reinforcement learning (RL) has been widely used in different research fields ranging from psychology to computer science. The unfeasibility of sampling all possibilities for continuous-state problems and the absence of an explicit teacher make RL algorithms preferable for supervised learning in the machine learning area, as the optimal control problem has become a popular su...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: Applied Intelligence

سال: 2022

ISSN: ['0924-669X', '1573-7497']

DOI: https://doi.org/10.1007/s10489-022-03191-2