Deep Reinforcement Learning for Instruction Following Visual Navigation in 3D Maze-Like Environments

Devo, Alessandro; Valigi, Paolo

doi:10.1109/lra.2020.2965857

Cited by 29 publications

(15 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition to using a first-view image to express the target location, Devo et al [53] studied the navigation task that follows natural language instruction input. Hsu et al [54] divided the complex indoor environment into different local areas, and generated navigation actions based on the scene image and target location.…”

Section: Developmentmentioning

confidence: 99%

Deep reinforcement learning based mobile robot navigation: A review

Zhu

Zhang

2021

Tsinghua Sci. Technol.

256

View full text Add to dashboard Cite

Navigation is a fundamental problem of mobile robots, for which Deep Reinforcement Learning (DRL) has received significant attention because of its strong representation and experience learning abilities. There is a growing trend of applying DRL to mobile robot navigation. In this paper, we review DRL methods and DRL-based navigation frameworks. Then we systematically compare and analyze the relationship and differences between four typical application scenarios: local obstacle avoidance, indoor navigation, multi-robot navigation, and social navigation. Next, we describe the development of DRL-based navigation. Last, we discuss the challenges and some possible solutions regarding DRL-based navigation.

show abstract

Section: Developmentmentioning

confidence: 99%

Deep reinforcement learning based mobile robot navigation: A review

Zhu

Zhang

2021

Tsinghua Sci. Technol.

256

View full text Add to dashboard Cite

show abstract

“…The mentioned VLN algorithms assume that objects in the environment, such as offices or houses, can be formulated into instructions. Different from these VLNs, Devo et al [115] focused on situations where objects in the environment cannot be specified as a navigation path, and considered 3D mazelike environments as the test bench which are very large and offer very intricate structures. This new VLN architecture can explicitly interpret the instructions and understands the direction to take along the path to navigate the environment without reference points.…”

Section: E Vision-and-language Drl Vnavigationmentioning

confidence: 99%

A Survey on Visual Navigation for Artificial Agents With Deep Reinforcement Learning

Zeng

Wang

2020

IEEE Access

View full text Add to dashboard Cite

Visual navigation (vNavigation) is a key and fundamental technology for artificial agents' interaction with the environment to achieve advanced behaviors. Visual navigation for artificial agents with deep reinforcement learning (DRL) is a new research hotspot in artificial intelligence and robotics that incorporates the decision making of DRL into visual navigation. Visual navigation via DRL, an end-to-end method, directly receives the high-dimensional images and generates an optimal navigation policy. In this paper, we first present an overview on reinforcement learning (RL), deep learning (DL) and deep reinforcement learning (DRL). Then, we systematically describe five main categories of visual DRL navigation: direct DRL vNavigation, hierarchical DRL vNavigation, multi-task DRL vNavigation, memory-inference DRL vNavigation and vision-language DRL vNavigation. These visual DRL navigation algorithms are reviewed in detail. Finally, we discuss the challenges and some possible opportunities to visual DRL navigation for artificial agents.

show abstract

“…Recent deep reinforcement learning (DRL) [ 7 ], which combines deep convolutional neural networks (CNNs) and reinforcement learning (RL) [ 8 ], provides a framework for learning control policy for specific tasks. DRL has achieved impressive results in many robotic tasks [ 9 – 11 ], including methods that attempt to finish autonomous exploration from raw sensory input.…”

Section: Introductionmentioning

confidence: 99%

End‐to‐End Autonomous Exploration with Deep Reinforcement Learning and Intrinsic Motivation

Ruan

Zhu

et al. 2021

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

Developing artificial intelligence (AI) agents is challenging for efficient exploration in visually rich and complex environments. In this study, we formulate the exploration question as a reinforcement learning problem and rely on intrinsic motivation to guide exploration behavior. Such intrinsic motivation is driven by curiosity and is calculated based on episode memory. To distribute the intrinsic motivation, we use a count-based method and temporal distance to generate it synchronously. We tested our approach in 3D maze-like environments and validated its performance in exploration tasks through extensive experiments. The experimental results show that our agent can learn exploration ability from raw sensory input and accomplish autonomous exploration across different mazes. In addition, the learned policy is not biased by stochastic objects. We also analyze the effects of different training methods and driving forces on exploration policy.

show abstract

Deep Reinforcement Learning for Instruction Following Visual Navigation in 3D Maze-Like Environments

Cited by 29 publications

References 17 publications

Deep reinforcement learning based mobile robot navigation: A review

Deep reinforcement learning based mobile robot navigation: A review

A Survey on Visual Navigation for Artificial Agents With Deep Reinforcement Learning

End‐to‐End Autonomous Exploration with Deep Reinforcement Learning and Intrinsic Motivation

Contact Info

Product

Resources

About