Bench2Drive-Robust: Benchmarking Closed-Loop Autonomous Driving under Deployment Perturbations

• URL: http://arxiv.org/abs/2605.18059v1
• Date: Mon, 18 May 2026 08:45:24 GMT
• Title: Bench2Drive-Robust: Benchmarking Closed-Loop Autonomous Driving under Deployment Perturbations
• Abstract summary: We present Bench2Drive-Robust, to our knowledge the first device-centric robustness benchmark for closed-loop end-to-end autonomous driving.<n>We systematically evaluate deployment-oriented perturbations arising from three major sources: camera-stream failures, ego-state estimation errors, and compute-induced control delay.<n>Our results show that these deployment-related perturbations can substantially degrade closed-loop driving performance.
• Score: 122.22391796628408
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Robustness is a critical requirement for deploying autonomous driving systems in the real world. Existing robustness benchmarks for autonomous driving have made important progress in studying the effects of image-level corruptions, such as adverse weather or camera degradation, on perception modules and open-loop planning outputs. However, deployment can also involve system-level imperfections, such as inference latency and ego-state estimation errors, which remain less studied in closed-loop E2E-AD evaluation. These imperfections can accumulate through the feedback loop and destabilize control. In this work, we present Bench2Drive-Robust, to our knowledge the first device-centric robustness benchmark for closed-loop end-to-end autonomous driving under realistic deployment perturbations. We systematically evaluate deployment-oriented perturbations arising from three major sources: camera-stream failures (frame drop, partial observation), ego-state estimation errors (GPS noise, and speed or odometry errors), and compute-induced control delay (model inference delay). We evaluate representative end-to-end driving methods and analyze their robustness under different perturbation severities. Our results show that these deployment-related perturbations can substantially degrade closed-loop driving performance, revealing robustness challenges that are not fully captured by conventional image-level corruption evaluations. By establishing a closed-loop evaluation protocol and demonstrating the substantial impact of these deployment-oriented perturbations, Bench2Drive-Robust defines practical robustness problems for end-to-end autonomous driving and encourages further research on deployment-aware robust driving systems.

Related papers

• A Multi-Modal Perception Pipeline for Object Detection and Tracking in Autonomous Racing [1.3281177137699656]
  This paper presents a multi-modal late-fusion perception pipeline for object detection and tracking in an autonomous racing domain.<n>Independent detections from cameras, LiDARs, and RADARs are combined to provide timely and robust state estimates of surrounding vehicles.<n>The tracking method explicitly compensates for detection delays and embeds in its model prior knowledge of vehicle dynamics and track layout.
  arXiv  Detail & Related papers  (2026-09-08T07:11:38Z)
• End-to-end Conditional Diffusion for Realistic and Controllable Visual Traffic Scenario Generation [59.188006170485046]
  E2E-CDiff is an end-to-end conditional diffusion framework for controllable and realistic scenario generation.<n>Differentiable guidance regulates speed, enforces drivable-area compliance, and supports collision-avoidance or collision-seeking behaviors.<n>E2E-CDiff also performs competitively as a learning-based ego planner.
  arXiv  Detail & Related papers  (2026-07-21T02:13:25Z)
• Real-World Perturbation Testing of Autonomous Driving Systems [0.9956658791307307]
  We present a large-scale study of 72 camera and LiDAR perturbations, evaluated across three testing modalities.<n>For camera-based systems, perturbations with limited offline impact can still induce unstable control and failures in real-world driving.<n>We show that real-time feasibility is a key constraint in real-world testing.
  arXiv  Detail & Related papers  (2026-07-06T11:34:31Z)
• RACF: A Resilient Autonomous Car Framework with Object Distance Correction [4.626261940793026]
  We present a Resilient Autonomous Car Framework (RACF) that incorporates an Object Distance Correction Algorithm (ODCA)<n>In this framework, when obstacle distance estimation produced by depth camera is inconsistent, a cross-sensor gate activates the correction algorithm to fix the detected inconsistency.<n>The framework achieved up to 35% RMSE reduction under strong corruption and improves stop compliance and braking latency, while operating in real time.
  arXiv  Detail & Related papers  (2026-04-14T08:06:09Z)
• When Sensors Fail: Temporal Sequence Models for Robust PPO under Sensor Drift [64.37959940809633]
  We study robustness of Proximal Policy Optimization (PPO) under temporally persistent sensor failures.<n>We show Transformer-based sequence policies substantially outperform, RNN, and SSMs in robustness, maintaining high returns even when large fractions of sensors are unavailable.
  arXiv  Detail & Related papers  (2026-03-04T22:21:54Z)
• Fixed-time control with prescribed performance for path following of underwater gliders [1.3149034455953847]
  This paper presents a fixed-time prescribed performance control scheme for the 3D path following of underwater gliders.<n>The primary contribution is the integration of a finite-time performance function within a fixed-time control framework.<n>A second key contribution is the development of a fixed-time sliding mode disturbance observer.
  arXiv  Detail & Related papers  (2025-12-23T20:11:25Z)
• Sequence of Expert: Boosting Imitation Planners for Autonomous Driving through Temporal Alternation [12.450883696383878]
  Imitation learning (IL) has emerged as a central paradigm in autonomous driving.<n>IL excels in matching expert behavior in open-loop settings by minimizing per-step prediction errors.<n>Over successive planning cycles, small, often imperceptible errors compound, potentially resulting in severe failures.<n>We propose Sequence of Experts (SoE) to enhance closed-loop performance without increasing model size or data requirements.
  arXiv  Detail & Related papers  (2025-12-15T08:50:23Z)
• ADPerf: Investigating and Testing Performance in Autonomous Driving Systems [7.10115369804893]
  We present the first comprehensive investigation on measuring and modeling the performance of the obstacle detection modules in two industry-grade autonomous driving systems.<n>We introduce ADPerf, a tool that aims to generate realistic point cloud data test cases that can expose increased detection latency.<n>Our evaluation highlights the need to conduct performance testing of obstacle detection components, especially 3D obstacle detection.
  arXiv  Detail & Related papers  (2025-10-15T01:45:59Z)
• DriveTransformer: Unified Transformer for Scalable End-to-End Autonomous Driving [62.62464518137153]
  DriveTransformer is a simplified E2E-AD framework for the ease of scaling up.<n>It is composed of three unified operations: task self-attention, sensor cross-attention, temporal cross-attention.<n>It achieves state-of-the-art performance in both simulated closed-loop benchmark Bench2Drive and real world open-loop benchmark nuScenes with high FPS.
  arXiv  Detail & Related papers  (2025-03-07T11:41:18Z)
• DRIVE: Dependable Robust Interpretable Visionary Ensemble Framework in Autonomous Driving [1.4104119587524289]
  Recent advancements in autonomous driving have seen a paradigm shift towards end-to-end learning paradigms. These models often sacrifice interpretability, posing significant challenges to trust, safety, and regulatory compliance. We introduce DRIVE, a comprehensive framework designed to improve the dependability and stability of explanations in end-to-end unsupervised driving models.
  arXiv  Detail & Related papers  (2024-09-16T14:40:47Z)
• Exploring the Causality of End-to-End Autonomous Driving [57.631400236930375]
  We propose a comprehensive approach to explore and analyze the causality of end-to-end autonomous driving. Our work is the first to unveil the mystery of end-to-end autonomous driving and turn the black box into a white one.
  arXiv  Detail & Related papers  (2024-07-09T04:56:11Z)
• Efficient and Robust LiDAR-Based End-to-End Navigation [132.52661670308606]
  We present an efficient and robust LiDAR-based end-to-end navigation framework. We propose Fast-LiDARNet that is based on sparse convolution kernel optimization and hardware-aware model design. We then propose Hybrid Evidential Fusion that directly estimates the uncertainty of the prediction from only a single forward pass.
  arXiv  Detail & Related papers  (2021-05-20T17:52:37Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.