计算机工程与应用2024，Vol.60Issue(17)：107-116,10.DOI:10.3778/j.issn.1002-8331.2306-0159
自动驾驶场景下的行人意图语义VSLAM
Pedestrian Intent Semantic VSLAM in Automatic Driving Scenarios
摘要
Abstract
Visual simultaneous localization and mapping(VSLAM)has found extensive applications in the field of auton-omous driving.However,conventional algorithms lack semantic information and are incapable of inferring or predicting pedestrians'behaviors or intentions within a scene.This paper introduces an effective semantic VSLAM method that employs a semantic segmentation algorithm based on dense prediction transformer(DPT)to acquire segmentation masks for potential dynamic targets,enabling dynamic feature removal.Given that the majority of dynamic objects in autono-mous driving scenarios are pedestrians and vehicles,in order to both reintegrate static points from potential dynamic tar-gets and re-detect dynamic objects,a geometric constraint is employed to jointly optimize camera poses while predicting pedestrian intentions.To accurately forecast whether pedestrians are crossing the road,a dual-stream,spatiotemporal adap-tive graph convolutional neural network is built using human skeletal information to predict pedestrian jaywalking inten-tions.The results validated on the KITTI dataset indicate that the proposed approach,in comparison to the ORB-SLAM3 algorithm,has a certain reduction in absolute trajectory estimation errors,demonstrating superior precision compared to algorithms of similar nature.This method holds the potential to furnish autonomous driving systems with richer semantic information,thereby enhancing the accomplishment of autonomous driving tasks.关键词
自动驾驶/语义分割/相机位姿优化/行人意图预测Key words
autonomous driving/semantic segmentation/camera pose optimization/pedestrian intention prediction分类
信息技术与安全科学引用本文复制引用
罗朝阳,张荣芬,刘宇红,李金,范润泽..自动驾驶场景下的行人意图语义VSLAM[J].计算机工程与应用,2024,60(17):107-116,10.基金项目
贵州省基础研究(自然科学)项目(黔科合基础-ZK[2021]重点001). （自然科学）
