[Paper] Isolation-aware Scheduling Framework for DNN-based End-to-End Autonomous Driving System on Tile-based Accelerators

Source: arXiv - 2606.10303v1

Overview

Level-4+ autonomous driving systems (ADS) must run dozens of heterogeneous deep neural networks (DNNs) as end-to-end (E2E) pipelines under a strict latency constraint (<=100 ms), even as execution time varies by up to 3.3x. Cost rules out dedicating isolated hardware to each function in mass-produced ADS, so these DNNs must be densely colocated on a single chip, which introduces shared-resource contention. Tile-based accelerators expose two scheduling opportunities that conventional ADS schedulers do not exploit. First, they provide a tunable degree of parallelism (DoP): assigning more tiles raises DoP and can shorten DNN execution time. Second, they provide hardware-native isolation: tiles can be physically partitioned among co-located DNNs. But using this flexibility is expensive: changing a task’s DoP triggers a stop-migrate-restart reallocation of its weights and intermediate features. At ADS task rates of 10-240 Hz, these stalls accumulate along E2E chains and threaten deadlines. Reservation-based schedulers fix DoP and leave this flexibility unused; work-conserving schedulers exploit it but assume reallocation is cheap and treat deadlines as independent. We present ADS-Tile that combines configurable isolation and elastic reservation into a spatio-temporal isolation-sharing space that bounds where and when reallocation occurs; a probabilistic latency model and a DAG-aware runtime scheduler then use this space to decide task colocation and DoP under shared E2E deadlines. On an industry- and academia- derived ADS benchmark, ADS-Tile uses up to 32% fewer tiles than the work-conserving baseline in deadline-critical settings and cuts reallocation-induced wasted processing capacity from 17%-44% to below 1.2%. Controlled spatio-temporal sharing improves resource efficiency and latency predictability for tile-based ADS.

Key Contributions

This paper presents research in the following areas:

• cs.AR
• cs.DC

Methodology

Please refer to the full paper for detailed methodology.

Practical Implications

This research contributes to the advancement of cs.AR.

Authors

• Chenguang Zhang
• Yuanpeng Zhang
• Chenhao Xue
• Yihan Yin
• Chen Zhang
• Guangyu Sun

Paper Information

• arXiv ID: 2606.10303v1
• Categories: cs.AR, cs.DC
• Published: June 9, 2026
• PDF: Download PDF