DEVA - Autonomous Driving with Emergent Visual Attention

DEVA - 具有紧急视觉注意力的自动驾驶

基本信息

批准号：
EP/N035399/1
负责人：
Nicolas Pugeault
金额：
$ 12.61万
依托单位：
UNIVERSITY OF EXETER
依托单位国家：
英国
项目类别：
Research Grant
财政年份：
2016
资助国家：
英国
起止时间：
2016 至无数据
项目状态：
已结题

来源：
https://gtr.ukri.org/projects?ref=EP%2FN035399%2F1
关键词：
DEVA Autonomous Driving Emergent Visual

项目摘要

How does a racer drive around a track? Approaching a bend in the road, a driver needs to monitor the road, steer around curves, manage speed and plan a trajectory avoiding collisions with other cars - and all of this, fast and accurately. For robots this remains a challenge: despite progress in computer vision over the last decades, artificial vision systems remain far from human vision in performance, robustness and speed. As a consequence, current prototypes of self-driving cars rely on a wide variety of sensors to palliate the limitations of their visual perception. One crucial aspect that distinguishes human from artificial vision is our capacity to focus and shift our attention. This project will propose a new model of visual attention for a robot driver, and investigate how attention focusing can be learnt automatically by trying to improve the robot's driving.How and where we focus our attention when solving a task such as driving is studied by psychologists, and the numerous models of attention can be sorted in two categories: first, top-down models capture how world knowledge and expectations guide our attention when performing a specific task; second, bottom-up models characterise how properties of the visual signal make specific regions capture our attention, a property often referred to as saliency. Yet, from a robotics perspective, there remains a lack of a unified framework describing the interplay of bottom-up and top-down attention, especially for a dynamic, time-critical task such as driving. In the racing scenario described above, the driver must take quick and decisive action to steer around bends and avoid obstacles - efficient use of attention is therefore critical.This project will investigate the hypothesis that our attention mechanisms are learnt on a task specific basis, in a such a way as to provide our visual system optimal information for performing the task. We will investigate how state-of-the-art computer vision and machine learning approaches can be used to learn attention, perception and action jointly to allow a robot driver to compete with humans on a racing simulator, using visual perception only.A generic learning framework for task-specific attention will be developed that is applicable across a broad range of visual tasks, and bearsthe potential for reducing the gap with human performance by a critical reduction in current processing times.

赛车手如何绕过赛道？在道路上弯曲时，驾驶员需要监视道路，绕曲线转动，管理速度并计划轨迹，以避免与其他汽车发生碰撞 - 以及所有这些，快速准确。对于机器人来说，这仍然是一个挑战：尽管过去几十年来计算机视觉的进展，人造视觉系统远非人类的性能，稳健性和速度的视野。结果，当前的自动驾驶汽车原型依靠多种传感器来抑制其视觉感知的局限性。将人与人造视力区分开来的一个关键方面是我们集中注意力和转移注意力的能力。该项目将为机器人驾驶员提出一种新的视觉关注模型，并研究如何通过尝试改善机器人的驾驶方式自动学习注意力集中。如何在解决诸如驾驶之类的任务（例如驾驶）时我们将注意力集中在心理学家中，并且可以通过心理学家进行研究，众多关注模型可以分为两类：首先，自上而下的，自上而下的模型指导我们在执行特定特定任务时，请注意我们的注意力；其次，自下而上的模型表征了视觉信号的特性如何使特定区域吸引我们的注意力，这通常称为显着性。然而，从机器人的角度来看，仍然缺乏描述自下而上和自上而下注意力的相互作用的统一框架，尤其是对于诸如驾驶等动态，关键时期的任务。在上面描述的赛车场景中，驾驶员必须采取快速而决定性的行动来引导弯曲并避免障碍 - 因此，有效使用注意力至关重要。该项目将调查以下假设：我们的注意机制是以特定于任务的基础学习的，以一种为执行任务提供视觉系统的方式。我们将调查如何使用仅使用视觉感知来允许机器人驾驶员在赛车模拟器上与人类竞争，以允许机器人驾驶员在赛车上与人竞争，从而开发出特定于任务的关注的通用学习框架将在范围内进行广泛的任务范围内，并且可以通过人类的范围来重塑，这将是适用的。

项目成果

期刊论文数量（9）

专著数量（0）

科研奖励数量（0）

会议论文数量（0）

专利数量（0）

Continuous Control With a Combination of Supervised and Reinforcement Learning

DOI：
10.1109/ijcnn.2018.8489702
发表时间：
2018-07
期刊：
2018 International Joint Conference on Neural Networks (IJCNN)
影响因子：
0
作者：
D. Kangin;N. Pugeault
通讯作者：
D. Kangin;N. Pugeault

Understanding and Visualizing Deep Visual Saliency Models

DOI：
10.1109/cvpr.2019.01045
发表时间：
2019-01-01
期刊：
2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2019)
影响因子：
0
作者：
He, Sen;Tavakoli, Hamed R.;Pugeault, Nicolas
通讯作者：
Pugeault, Nicolas

Aggregated Sparse Attention for Steering Angle Prediction