.png){kind=icon}
返回
Grounded Relational Inference: Domain Knowledge Driven Explainable Autonomous Driving
.png){kind=icon}
OA
AI
DOI：10.1109/TITS.2024.3424667.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Explainability is essential for autonomous vehicles and other robotics systems interacting with humans and other objects during operation. Humans need to understand and anticipate the actions taken by machines for trustful and safe cooperation. In this work, we aim to develop an explainable model that generates explanations consistent with both human domain knowledge and the model's inherent causal relation. In particular, we focus on an essential building block of autonomous drivingmulti-agent interaction modeling. We propose Grounded Relational Inference (GRI). It models an interactive system's underlying dynamics by inferring an interaction graph representing the agents' relations. We ensure a semantically meaningful interaction graph by grounding the relational latent space into semantic interactive behaviors defined with expert domain knowledge. We demonstrate that it can model interactive traffic scenarios under both simulation and real-world settings and generate semantic graphs explaining the vehicle's behavior through their interactions.
Keyword:
Atmospheric modeling
Semantics
Autonomous vehicles
Trajectory
Graph neural networks
Decoding
Aerospace electronics
explainable AI
deep learning
driving behavior modeling
期刊
IF:
8.4
论文数:
9.5K
被引数:
6.3W
机构
引用论文
Genetic Relationships Between the G Protein βγ Complex, Ste5p, Ste20p and Cdc42p: Investigation of Effector Roles in the Yeast Pheromone Response Pathway.png)
Genetics
OAAI
GeneticsIF0
OAAI分享收藏Neuropsychological assessment of African children: evidence for a universal brain/behavior omnibus within a coconstructivist paradigm非洲儿童的神经心理学评估: 共同建构主义范式中普遍的大脑/行为综合的证据
PREAI
分享收藏
查看更多内容