analiza aiqbench
This research presents a CPU-optimized LiDAR pipeline using Random Forests that achieves near-perfect cone detection latency, offering a viable alternative to GPU-dependent deep learning models for resource-constrained autonomous racing applications. By reducing feature inputs and eliminating heavy neural networks, the study provides a reproducible baseline specifically designed for edge computing environments where power and processing limits are strict.
Lightweight LiDAR-Based Cone Detection Framework Using Random Forest for Formula Student Driverless
arXiv:2609.11527v1 Announce Type: new Abstract: Reliable, low-latency perception is crucial for Formula Student Driverless vehicles, yet many existing pipelines rely on deep learning and multi-sensor fusion, often requiring GPU acceleration. This paper presents a lightweight LiDAR-only perception pipeline tailored for CPU execution, combining ground removal, IMU-based motion compensation, DBSCAN clustering, and geometric feature-based Random Forest classification. Feature importance analysis reduced the model input from 12 to 7 features while preserving performance. Evaluated on 2,371 labeled clusters collected from real FSD events, the pipeline achieves an F1-score of 98.33% and an end-to-end runtime of 3.13 ms on CPU-only hardware. The released dataset, labeling tool, and trained models provide a practical and reproducible baseline for other resource-constrained autonomous racing teams.
Abstract:Reliable, low-latency perception is crucial for Formula Student Driverless vehicles, yet many existing pipelines rely on deep learning and multi-sensor fusion, often requiring GPU acceleration. This paper presents a lightweight LiDAR-only perception pipeline tailored for CPU execution, combining ground removal, IMU-based motion compensation, DBSCAN clustering, and geometric feature-based Random Forest classification. Feature importance analysis reduced the model input from 12 to 7 features while preserving performance. Evaluated on 2,371 labeled clusters collected from real FSD events, the pipeline achieves an F1-score of 98.33% and an end-to-end runtime of 3.13 ms on CPU-only hardware. The released dataset, labeling tool, and trained models provide a practical and reproducible baseline for other resource-constrained autonomous racing teams.
9 pages, 2 figures, 3 tables. Accepted at the 5th International Conference on Cognitive Mobility (CogMob 2026)
Artificial Intelligence (cs.AI)
arXiv:2609.11527 [cs.AI]
(or arXiv:2609.11527v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2609.11527 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 224 cuvinte
Subiecte conexe
RoughSense: Lightweight Terrain-Induced Rover Vibration Prediction Using Point Clouds and IMU Feedback
arXiv · 4 sept. 2026
Lightweight 3D Object Detection via Mamba-Based Knowledge Distillation
arXiv · 5 aug. 2026
TACTIC: Temporal and Context-Aware LLM Tactical Planning for Roadside LiDAR Attacks
arXiv · 1 oct. 2026
JEPLO: Joint-Embedding Predictive Learning for LiDAR-Based Legged Locomotion
arXiv · 15 sept. 2026