摘要
针对深度强化学习视觉导航算法因导航场景变化而导致导航精度下降,影像匹配的实时性和可靠性降低的问题,提出一种融合拆分注意力机制和下一次预期观测(NEO)的视觉导航模型。首先使用ResNest50骨干网络提取当前状态和目标状态的特征以降低网络冗余,利用跨阶段部分连接CSP强化捕获浅层目标特征信息以增强模型的学习能力。然后提出改进的损失函数,使得推理网络更加接近于真实后验,从而智能体能在当前环境下做出最佳决策,进一步提升不同场景下模型的导航精度。在AVD数据集和AI2-THOR场景进行训练及测试,实验结果表明,本文算法导航精度高达96.8%,平均SR提升约3%,平均SPL提升约6%,可以满足导航精度和实时匹配的要求。
A visual navigation model incorporating split attention mechanism and next expected observation(NEO)is proposed to address the problem that deep reinforcement learning visual navigation algorithm degrades navigation accuracy,real-time and reliability of image matching due to navigation scene changes.The features of current and target states are first extracted using the ResNest50 backbone network to reduce network redundancy.The shallow target feature information is captured intensively using a cross-stage-partial-connections CSP to enhance the learning ability of the model.Then an improved loss function is proposed to make the inference network closer to the true posterior so that the agent can make the best decision in the current environment and further improve the navigation accuracy of the model in different scenarios.The training and testing are conducted on AVD dataset and AI2-THOR scenes,and the experimental results show that the navigation accuracy of the algorithm in this paper is as high as 96.8%,with an average SR improvement of about 3%and an average SPL improvement of about 6%,which meets the requirements of navigation accuracy and real-time matching.
作者
刘紫燕
杨模
袁浩
梁静
梁水波
孙昊堃
Liu Ziyan;Yang Mo;Yuan Hao;Liang Jing;Liang Shuibo;Sun Haokun(College of Big Data and Information Engineering,Guizhou University,Guiyang 550025,China;State Key Laboratory of Public Big Data,Guizhou University,Guiyang 550025,China)
出处
《电子测量与仪器学报》
CSCD
北大核心
2023年第1期96-105,共10页
Journal of Electronic Measurement and Instrumentation
基金
贵州省科学技术基金(黔科合基础[2016]1054)
贵州省联合资金(黔科合LH字[2017]7226号)
贵州大学2017年度学术新苗培养及创新探索专项(黔科合平台人才[2017]5788)
贵州省科技计划项目(黔科合SY字[2011]3111)资助
关键词
视觉导航
深度强化学习
拆分注意力机制
下一次预期观测
visual navigation
deep reinforcement learning
split attention mechanism
next expected observations(NEO)
