Meta-IRLSOT++: A meta-inverse reinforcement learning method for fast adaptation of trajectory prediction networks

计算机科学 弹道 强化学习 人工智能 机器学习 天文 物理
作者
Biao Yang,Yanan Lu,Rui Wan,Hongyu Hu,Changchun Yang,Rongrong Ni
出处
期刊:Expert Systems With Applications [Elsevier BV]
卷期号:240: 122499-122499 被引量:46
标识
DOI:10.1016/j.eswa.2023.122499
摘要

Recent research on pedestrian trajectory prediction based on deep learning has made significant progress. However, the previous methods do not deeply explore the relationship between scene information and trajectory. Moreover, the training model requires massive data and only targets specific scenes, so the prediction performance is poor when the new scene samples are limited. To solve the above problems, a meta-inverse reinforcement learning-based framework, dubbed Meta-IRLSOT++, is proposed in this work. IRLSOT++ improves the solid baseline IRLSOT, and the main contributions are as follows: (1) An inverse reinforcement learning framework is introduced to explore the trajectory–scene association to achieve task-level scene understanding, enhancing the correlation between trajectory and scene. The trajectory heat maps describing pedestrians dynamic characteristics are leveraged to align better the trajectory and scene semantic segmentation predicted by TopFormer. (2) A Transformer-based encoder–decoder network is proposed to fuse the trajectory and plan cues for better generating multi-modal trajectories with multi-head attention. The social graph attention and pseudo-oracle predictor are introduced to capture pedestrians’ social interactions and intent states, ameliorating trajectory prediction performance. (3) Meta-learning is leveraged to achieve collaborative training based on IRLSOT++ to improve the model generalization in new scenes, enabling fast adaptation of trajectory prediction. Experimental results on the Stanford Drone Dataset (SDD) indicate that IRLSOT++ can precisely forecast future trajectories by improving IRLSOT, decreasing Average Displacement Error/ Final Displacement Error (ADE/FDE) values from 9.66/13.05 to 8.36/12.28. Moreover, the meta-learning strategy quickly adapts IRLSOT++ to the new scene, achieving ADE/FDE values of 7.31/11.02 and 15.42/26.55 when Tpred is set to 12 and 24, respectively. Both quantitative and qualitative experimental results show that Meta-IRLSOT++ has accuracy and fast adaptability, which is beneficial for real-world trajectory prediction tasks.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
szy完成签到,获得积分0
刚刚
自由月亮完成签到,获得积分10
1秒前
LYZSh完成签到,获得积分10
2秒前
科研通AI6.2的应助被茴香采纳,获得10
2秒前
DND完成签到,获得积分10
3秒前
butter完成签到,获得积分10
3秒前
5秒前
5秒前
科目三的应助被Sapphire采纳,获得30
7秒前
圆圈完成签到,获得积分10
9秒前
ding的应助被CC采纳,获得10
9秒前
10秒前
qiu完成签到 ,获得积分10
10秒前
KissesU完成签到 ,获得积分10
10秒前
HaohaoLi完成签到,获得积分10
10秒前
11秒前
勺子爱西瓜完成签到,获得积分0
12秒前
麻团儿发布了新的文献求助10
12秒前
zain完成签到 ,获得积分10
13秒前
赫凯完成签到,获得积分10
15秒前
慢慢完成签到 ,获得积分10
15秒前
mei完成签到,获得积分10
16秒前
18秒前
sxd完成签到 ,获得积分10
18秒前
19秒前
英俊的铭的应助被科研通管家采纳,获得10
19秒前
科目三的应助被科研通管家采纳,获得30
19秒前
共享精神的应助被科研通管家采纳,获得10
20秒前
情怀的应助被科研通管家采纳,获得10
20秒前
脑洞疼的应助被科研通管家采纳,获得10
20秒前
orixero的应助被科研通管家采纳,获得30
20秒前
小二郎的应助被科研通管家采纳,获得10
20秒前
20秒前
烟花的应助被科研通管家采纳,获得10
20秒前
所所的应助被科研通管家采纳,获得10
20秒前
英俊的铭的应助被科研通管家采纳,获得10
21秒前
bkagyin的应助被科研通管家采纳,获得10
21秒前
jia完成签到,获得积分10
21秒前
22秒前
Grubby123发布了新的文献求助30
23秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1700
The World Atlas of Wine 9th edition 1000
奈特基础组织学:配套组织病理学(第4版)(Netter's Essential Histology With Correlated Histopathology) 800
八旗制度の研究 600
高分子基功能界面水蒸发材料与应用研究 500
外场增强的水汽发电研究 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8185571
求助须知:如何正确求助?哪些是违规求助? 9574336
关于积分的说明 21835187
捐赠科研通 7783367
什么是DOI,文献DOI怎么找? 3434606
关于科研通互助平台的介绍 2558985
邀请新用户注册赠送积分活动 2481374