Edge Intelligence for Autonomous Driving in 6G Wireless System: Design Challenges and Solutions

Abstract

In a level-5 autonomous driving system, the autonomous driving vehicles (AVs) are expected to sense the surroundings via analyzing a large amount of data captured by a variety of onboard sensors in near-real-time. As a result, enormous computing costs will be introduced to the AVs for processing the tasks with the deployed machine learning (ML) model, while the inference accuracy may not be guaranteed. In this context, the advent of edge intelligence (EI) and sixthgeneration (6G) wireless networking are expected to pave the way to more reliable and safer autonomous driving by providing multi-access edge computing (MEC) together with ML to AVs in close proximity. To realize this goal, we propose a two-tier EIempowered autonomous driving framework. In the autonomousvehicles tier, the autonomous vehicles are deployed with the shallow layers by splitting the trained deep neural network model. In the edge-intelligence tier, an edge server is implemented with the remaining layers (also deep layers) and an appropriately trained multi-task learning (MTL) model. In particular, obtaining the optimal offloading strategy (including the binary offloading decision and the computational resources allocation) can be formulated as a mixed-integer nonlinear programming (MINLP) problem, which is solved via MTL in near-real-time with high accuracy. On another note, an edge-vehicle joint inference is proposed through neural network segmentation to achieve efficient online inference with data privacy-preserving and less communication delay. Experiments demonstrate the effectiveness of the proposed framework, and open research topics are finally listed.

Amalgamation of blockchain and sixth‐generation‐envisioned responsive edge orchestration in future cellular vehicle‐to‐anything ecosystems: Opportunities and challenges

“…Thus, the integration can support mission-critical operations in C-V2X and can allow local sensor objects to interact with other devices with reduced operational complexity. Through AI, it can support operations that offer real-time and on-demand services to CASVs through the 6G communication channels, F I G U R E 7 Sixth-generation-assisted intelligent and responsive edge 35 minimizing road accidents. We present two of the possible use-cases that benefit from the possible integration.…”

“…It includes orbital angular momentum (OAM) multiplexing, BC-based spectrum sharing mechanism, super-massive MIMO (SM-MIMO), and molecular communication. Yang et al 35 proposed a four-layer novel AI-based framework 6G networks for better service provisioning through which network adjustment is carried out automatically. Authors in Reference 37 performed the comparative analysis of cellular V2X and 802.11 V2X and provide a concluding remark to excel in the performance of the IoV network.…”

“…The scheme measures the handover probability and reduces the latency of the UAV network from ground control. Yang et al 35 proposed object detection for CASV based on optimized offloading from base station. A multiclass classification problem is presented for the same, and edge resource allocation is considered as regression with mean square error training.…”

“…Based on changing C-V2X conditions, the model segments itself to provide optimized solutions to the neighboring cells. 35…”