Differentiable Integrated Motion Prediction and Planning With Learnable Cost Function for Autonomous Driving

可微函数 运动规划 计算机科学 稳健性(进化) 基线(sea) 弹道 规划师 功能(生物学) 场景测试 人工智能 控制理论(社会学) 机器学习 控制工程 控制(管理) 工程类 多样性(控制论) 机器人 数学 数学分析 生物化学 化学 海洋学 物理 天文 进化生物学 生物 基因 地质学
作者
Zhiyu Huang,Haochen Liu,Jingda Wu,Chen Lv
出处
期刊:IEEE transactions on neural networks and learning systems [Institute of Electrical and Electronics Engineers]
卷期号:35 (11): 15222-15236 被引量:39
标识
DOI:10.1109/tnnls.2023.3283542
摘要

Predicting the future states of surrounding traffic participants and planning a safe, smooth, and socially compliant trajectory accordingly are crucial for autonomous vehicles (AVs). There are two major issues with the current autonomous driving system: the prediction module is often separated from the planning module, and the cost function for planning is hard to specify and tune. To tackle these issues, we propose a differentiable integrated prediction and planning (DIPP) framework that can also learn the cost function from data. Specifically, our framework uses a differentiable nonlinear optimizer as the motion planner, which takes as input the predicted trajectories of surrounding agents given by the neural network and optimizes the trajectory for the AV, enabling all operations to be differentiable, including the cost function weights. The proposed framework is trained on a large-scale real-world driving dataset to imitate human driving trajectories in the entire driving scene and validated in both open-loop and closed-loop manners. The open-loop testing results reveal that the proposed method outperforms the baseline methods across a variety of metrics and delivers planning-centric prediction results, allowing the planning module to output trajectories close to those of human drivers. In closed-loop testing, the proposed method outperforms various baseline methods, showing the ability to handle complex urban driving scenarios and robustness against the distributional shift. Importantly, we find that joint training of planning and prediction modules achieves better performance than planning with a separate trained prediction module in both open-loop and closed-loop tests. Moreover, the ablation study indicates that the learnable components in the framework are essential to ensure planning stability and performance. Code and Supplementary Videos are available at https://mczhi.github.io/DIPP/.

科研通智能强力驱动
Strongly Powered by AbleSci AI
更新
大幅提高文件上传限制,最高150M (2024-4-1)

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
1秒前
大胆的夏天完成签到,获得积分10
2秒前
2秒前
shijin发布了新的文献求助10
3秒前
李健的小迷弟应助微糖采纳,获得10
5秒前
阔达的蜜粉完成签到,获得积分10
5秒前
5秒前
5秒前
yee完成签到,获得积分10
5秒前
打打应助少一点丶天分采纳,获得10
6秒前
8秒前
万能图书馆应助小可爱采纳,获得30
9秒前
科研通AI2S应助阔达的蜜粉采纳,获得10
10秒前
tuanheqi应助Jzt采纳,获得50
10秒前
10秒前
12秒前
12秒前
nyzcc完成签到,获得积分10
12秒前
12秒前
fagfagsf发布了新的文献求助10
13秒前
Singularity应助计算机小咖采纳,获得10
13秒前
14秒前
大气可燕发布了新的文献求助10
14秒前
建成完成签到,获得积分10
15秒前
敏感沛春发布了新的文献求助10
15秒前
机灵的仙人掌完成签到,获得积分10
16秒前
16秒前
16秒前
微糖发布了新的文献求助10
16秒前
18秒前
田様应助asa采纳,获得10
19秒前
20秒前
田様应助坦率的棉花糖采纳,获得10
20秒前
21秒前
nini发布了新的文献求助10
21秒前
可爱香芦完成签到 ,获得积分10
22秒前
魔幻的千山完成签到,获得积分10
22秒前
wind完成签到,获得积分10
22秒前
25秒前
25秒前
高分求助中
歯科矯正学 第7版(或第5版) 1004
The late Devonian Standard Conodont Zonation 1000
Nickel superalloy market size, share, growth, trends, and forecast 2023-2030 1000
Semiconductor Process Reliability in Practice 1000
Smart but Scattered: The Revolutionary Executive Skills Approach to Helping Kids Reach Their Potential (第二版) 1000
PraxisRatgeber: Mantiden: Faszinierende Lauerjäger 700
A new species of Coccus (Homoptera: Coccoidea) from Malawi 500
热门求助领域 (近24小时)
化学 医学 生物 材料科学 工程类 有机化学 生物化学 物理 内科学 纳米技术 计算机科学 化学工程 复合材料 基因 遗传学 催化作用 物理化学 免疫学 量子力学 细胞生物学
热门帖子
关注 科研通微信公众号,转发送积分 3238154
求助须知:如何正确求助?哪些是违规求助? 2883512
关于积分的说明 8230736
捐赠科研通 2551616
什么是DOI,文献DOI怎么找? 1380076
科研通“疑难数据库(出版商)”最低求助积分说明 648923
邀请新用户注册赠送积分活动 624589