# Performance Implications of Multi-Chiplet Neural Processing Units on Autonomous Driving Perception

arXiv: https://arxiv.org/abs/2411.16007 · HTML: https://mlsanity.com/paper/2411.16007/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 41 references that are in the map.
- First posted on arXiv 24 Nov 2024.

## Builds on (most cited first, 12 of 12)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [EfficientDet: Scalable and Efficient Object Detection](https://mlsanity.com/paper/1911.09070/) (2019)
- [Centaur: A Chiplet-based, Hybrid Sparse-Dense Accelerator for Personalized Recommendations](https://mlsanity.com/paper/2005.05968/) (2020)
- [HydraFusion: Context-Aware Selective Sensor Fusion for Robust and Efficient Autonomous Vehicle Perception](https://mlsanity.com/paper/2201.06644/) (2022)
- [Map-and-Conquer: Energy-Efficient Mapping of Dynamic Neural Nets onto Heterogeneous MPSoCs](https://mlsanity.com/paper/2302.12926/) (2023)
- [SAGE: A Split-Architecture Methodology for Efficient End-to-End Autonomous Vehicle Control](https://mlsanity.com/paper/2107.10895/) (2021)
- [DREAM: A Dynamic Scheduler for Dynamic Real-time Multi-model ML Workloads](https://mlsanity.com/paper/2212.03414/) (2022)
- [Gemini: Mapping and Architecture Co-exploration for Large-scale DNN Chiplet Accelerators](https://mlsanity.com/paper/2312.16436/) (2023)
- [H2H: Heterogeneous Model to Heterogeneous System Mapping with Computation and Communication Awareness](https://mlsanity.com/paper/2204.13852/) (2022)
- [SCAR: Scheduling Multi-Model AI Workloads on Heterogeneous Multi-Chiplet Module Accelerators](https://mlsanity.com/paper/2405.00790/) (2024)
- [Romanus: Robust Task Offloading in Modular Multi-Sensor Autonomous Driving Systems](https://mlsanity.com/paper/2207.08865/) (2022)
- [CARIn: Constraint-Aware and Responsive Inference on Heterogeneous Devices for Single- and Multi-DNN Workloads](https://mlsanity.com/paper/2409.01089/) (2024)

## Built on by (most cited first, 1 of 1)

- [Terastal: Layer-Variant-based Scheduling for Real-Time Multi-DNN Workloads on Heterogeneous Accelerators](https://mlsanity.com/paper/2606.06818/) (2026)

## Abstract (arXiv)

We study the application of emerging chiplet-based Neural Processing Units to accelerate vehicular AI perception workloads in constrained automotive settings. The motivation stems from how chiplets technology is becoming integral to emerging vehicular architectures, providing a cost-effective trade-off between performance, modularity, and customization; and from perception models being the most computationally demanding workloads in a autonomous driving system. Using the Tesla Autopilot perception pipeline as a case study, we first breakdown its constituent models and profile their performance on different chiplet accelerators. From the insights, we propose a novel scheduling strategy to efficiently deploy perception workloads on multi-chip AI accelerators. Our experiments using a standard DNN performance simulator, MAESTRO, show our approach realizes 82% and 2.8x increase in throughput and processing engines utilization compared to monolithic accelerator designs.