Paper Reading AI Learner

Plug-and-Play Sensor Evolution in Software-Defined Vehicles through Asset Administration Shells

2026-11-03 12:30:17
Akshay Narla, Nasser Jazdi, Michael Weyrich

Abstract

Software-defined vehicles (SDVs) promise continuous software evolution, but sensor integration in deployed vehicles still depends on manual configuration, hardware-specific drivers, and repeated application-side adaptation. Existing standards address only parts of this problem. AUTOSAR standardizes selected sensor interfaces, the Vehicle Signal Specification (VSS) standardizes signal semantics, and Asset Administration Shells (AAS) standardize digital asset descriptions. However, these building blocks are rarely combined into an end-to-end onboarding path for sensors attached after vehicle deployment. This paper presents an AAS-driven plug-and-play sensor integration framework for safety-supervised sensors. Each sensor is represented by an AAS that combines hardware identification, interface metadata, VSS mappings, and references to containerized driver and optional application software. After physical connection, an edge gateway identifies the device, retrieves the matching AAS, deploys the referenced software, and publishes the resulting data under stable vehicle signal paths. A prototype on a resource-constrained edge gateway connected to an RR100 UGV demonstrates automated integration of a USB camera and an ultrasonic sensor. Across repeated integration cycles, the prototype reaches operational state in about 2s, shows no failed events, and keeps applications independent of hardware-specific interfaces. The results indicate that the introduction of AAS as an executable onboarding contract enables automated runtime deployment of sensor drivers in SDV environments.

Abstract (translated)

URL

https://arxiv.org/abs/2611.23298

PDF

https://arxiv.org/pdf/2611.23298.pdf


Tags
3D Action Action_Localization Action_Recognition Activity Adversarial Agent Attention Autonomous Bert Boundary_Detection Caption Chat Classification CNN Compressive_Sensing Contour Contrastive_Learning Deep_Learning Denoising Detection Dialog Diffusion Drone Dynamic_Memory_Network Edge_Detection Embedding Embodied Emotion Enhancement Face Face_Detection Face_Recognition Facial_Landmark Few-Shot Gait_Recognition GAN Gaze_Estimation Gesture Gradient_Descent Handwriting Human_Parsing Image_Caption Image_Classification Image_Compression Image_Enhancement Image_Generation Image_Matting Image_Retrieval Inference Inpainting Intelligent_Chip Knowledge Knowledge_Graph Language_Model LLM Matching Medical Memory_Networks Multi_Modal Multi_Task NAS NMT Object_Detection Object_Tracking OCR Ontology Optical_Character Optical_Flow Optimization Person_Re-identification Point_Cloud Portrait_Generation Pose Pose_Estimation Prediction QA Quantitative Quantitative_Finance Quantization Re-identification Recognition Recommendation Reconstruction Regularization Reinforcement_Learning Relation Relation_Extraction Represenation Represenation_Learning Restoration Review RNN Robot Salient Scene_Classification Scene_Generation Scene_Parsing Scene_Text Segmentation Self-Supervised Semantic_Instance_Segmentation Semantic_Segmentation Semi_Global Semi_Supervised Sence_graph Sentiment Sentiment_Classification Sketch SLAM Sparse Speech Speech_Recognition Style_Transfer Summarization Super_Resolution Surveillance Survey Text_Classification Text_Generation Time_Series Tracking Transfer_Learning Transformer Unsupervised Video_Caption Video_Classification Video_Indexing Video_Prediction Video_Retrieval Visual_Relation VQA Weakly_Supervised Zero-Shot