analiza aiqbench
This research introduces Self-Supervised Imitation Learning (SSIL), a novel framework for end-to-end autonomous driving that eliminates the need for extensive human-labeled control data or pre-trained models. By leveraging LiDAR-derived vehicle poses to generate pseudo steering angles and employing cross-attention conditioning, SSIL achieves performance comparable to supervised learning benchmarks while addressing critical data scarcity issues in vision-based navigation systems.
SSIL: Self-Supervised Imitation Learning for End-to-End Driving
arXiv:2308.14329v4 Announce Type: replace Abstract: In autonomous driving, the end-to-end (E2E) driving approach that predicts vehicle control signals directly from sensor data is rapidly gaining attention. To learn a safe E2E driving system, one needs an extensive amount of driving data and human intervention. Vehicle control data is constructed by many hours of human driving, and it is challenging to construct large vehicle control datasets. Often, publicly available driving datasets are collected with limited driving scenes, and collecting vehicle control data is only available by vehicle manufacturers. To address these challenges, this paper proposes the first self-supervised learning framework, Self-Supervised Imitation Learning (SSIL), for E2E driving. The proposed SSIL framework can learn vision-based E2E driving networks without using driving command data or a pre-trained model. To construct pseudo steering angle data, proposed SSIL predicts a pseudo target from the vehicle's poses at the current and previous time points that are estimated with light detection and ranging sensors. In addition, we propose a new cross-attention-based conditioning approach (CACA) for a vision encoder in E2E driving, where a high-level instruction serves as the conditioning signal for visual information. Our numerical experiments with three different benchmark datasets demonstrate that the proposed SSIL framework achieves very comparable E2E driving accuracy with the supervised learning counterpart. Furthermore, the proposed pseudo-label predictor outperformed an existing one using proportional integral derivative controller, and proposed CACA achieved superior performance over existing conditioning approaches.
Abstract:In autonomous driving, the end-to-end (E2E) driving approach that predicts vehicle control signals directly from sensor data is rapidly gaining attention. To learn a safe E2E driving system, one needs an extensive amount of driving data and human intervention. Vehicle control data is constructed by many hours of human driving, and it is challenging to construct large vehicle control datasets. Often, publicly available driving datasets are collected with limited driving scenes, and collecting vehicle control data is only available by vehicle manufacturers. To address these challenges, this paper proposes the first self-supervised learning framework, Self-Supervised Imitation Learning (SSIL), for E2E driving. The proposed SSIL framework can learn vision-based E2E driving networks without using driving command data or a pre-trained model. To construct pseudo steering angle data, proposed SSIL predicts a pseudo target from the vehicle's poses at the current and previous time points that are estimated with light detection and ranging sensors. In addition, we propose a new cross-attention-based conditioning approach (CACA) for a vision encoder in E2E driving, where a high-level instruction serves as the conditioning signal for visual information. Our numerical experiments with three different benchmark datasets demonstrate that the proposed SSIL framework achieves very comparable E2E driving accuracy with the supervised learning counterpart. Furthermore, the proposed pseudo-label predictor outperformed an existing one using proportional integral derivative controller, and proposed CACA achieved superior performance over existing conditioning approaches.
Robotics (cs.RO); Artificial Intelligence (cs.AI)
arXiv:2308.14329 [cs.RO]
(or arXiv:2308.14329v4 [cs.RO] for this version)
Text integral sub licența deschisă a publicației · 243 cuvinte
Subiecte conexe
TerraZero: Procedural Driving Simulation for Zero-Demonstration Self-Play at Scale
arXiv · 6 aug. 2026
TerraTransfer: Learning End-to-End Driving Policies Without Expert Demonstrations
arXiv · 17 iun. 2026
Mastering Agentic Techniques: AI Agent Customization
NVIDIA · 20 mai 2026
Echoverse: Deep, evolving environments for computer-use agents
Microsoft Research · 30 iul. 2026