AESOP enables path-aware adversarial attacks that inflate FLOPs in ML pipelines by up to 2407x, 20x more than single-model baselines, even under defenses that force throughput collapse or data loss.
hub Canonical reference
Motion planning for autonomous driving: The state of the art and future perspectives
Canonical reference. 100% of citing Pith papers cite this work as background.
20
Pith papers citing it
118
external citations · OpenAlex
Background
100% of classified citations
hub tools
citation-role summary
background 6
citation-polarity summary
verdicts
UNVERDICTED 20roles
background 6polarities
background 6representative citing papers
A delay-compensating controller with revised formulation and brake-based stabilization enables a production combustion-engine car to sustain circular and figure-eight drifts with 1.1 m lateral error and 0.06 rad sideslip overshoot.
MC-Risk linearly composes motorized-agent, VRU, and road-penalty fields into a bird's-eye-view risk grid that achieves superior localization and early detection on RiskBench while serving as an MPC cost for risk-aware planning.
Driver-WM is a driver-centric latent world model for causal rollout of in-cabin dynamics conditioned on out-cabin traffic, unifying kinematics forecasting with behavioral and emotional recognition via dual-stream architecture and gated injection.
ESARBench is the first unified benchmark for MLLM-driven UAV agents that must explore, locate clues, and decide on victim positions in photorealistic simulated SAR environments.
Human car-following variability generates nonlinear damping via non-monotonic time-headway shifts that dissipates traffic perturbations, whereas commercial ACC amplifies them 2.7-5 fold in fuel use.
A distributed multi-UGV framework using descriptor-aided loop closure and loop-aware hierarchical planning reduces exploration time by 15% and travel distance by 14% versus mTSP baseline while achieving 89.9% AR@1 place recognition.
DiffSlack introduces learnable slack variables and a damped Gauss-Newton projection to create a differentiable layer that enforces hard nonlinear inequality constraints in neural network outputs.
Hyper-V2X uses a Bayesian hypernetwork with partial weight generation and V2X context embedding to produce calibrated epistemic and aleatoric uncertainty estimates for multi-agent BEV segmentation on the OPV2V benchmark.
Dr-BA delivers a separable optimization approach for direct radar bundle adjustment and cross-session localization using full spinning-radar intensity images, achieving state-of-the-art performance on over 200 km of on-road data.
EgoDyn-Bench finds a Perception Bottleneck: foundation models hold ego-motion logic in language but misalign it with vision, underperforming geometric baselines until given explicit trajectories.
LiloDriver uses LLMs and memory-augmented planning in a four-stage pipeline to outperform rule-based and learning-based methods on both common and rare scenarios in the nuPlan benchmark.
Recent LiDAR 3D detectors remain as vulnerable to adversarial attacks as predecessors, with voxel-based and non-anchor-based models showing greater susceptibility under a multi-factor robustness framework.
CooperScene provides 59K synchronized frames with 344K 3D annotations from multi-modal sensors on 3 CAVs and 1 RSU plus real C-V2X communication traces for cooperative autonomy benchmarking.
Proposes basing teleoperation ODD on minimal risk maneuver capability with a dedicated system, demonstrated via use case.
A framework integrates user language and probabilistic environment estimates into adaptive safety certificates that guarantee long-term safety for stochastic systems via probabilistic invariance.
Vision-language models can serve as zero-shot ODD sensors for autonomous driving when using definition-anchored chain-of-thought prompting with persona decomposition.
RadarCNN classifies indoor objects from radar IQ data at 97-99% accuracy, holding at ~50% under noise and occlusion.
A framework for multi-floor AGV trajectory planning that combines GVD-based task selection with optimization-based trajectory generation and constraint reduction, verified in simulation.
A survey that organizes methods for cross-domain object detection into a taxonomy, analyzes domain shift across detection stages, and outlines persistent challenges.
citing papers explorer
-
AESOP: Adversarial Execution-path Selection to Overload Deep Learning Pipelines
AESOP enables path-aware adversarial attacks that inflate FLOPs in ML pipelines by up to 2407x, 20x more than single-model baselines, even under defenses that force throughput collapse or data loss.
-
Drifting in the Future: Stabilizing Path Following Drifting on High-Latency Vehicle Systems
A delay-compensating controller with revised formulation and brake-based stabilization enables a production combustion-engine car to sustain circular and figure-eight drifts with 1.1 m lateral error and 0.06 rad sideslip overshoot.
-
MC-Risk: Multi-Component Risk Fields for Risk Identification and Motion Planning
MC-Risk linearly composes motorized-agent, VRU, and road-penalty fields into a bird's-eye-view risk grid that achieves superior localization and early detection on RiskBench while serving as an MPC cost for risk-aware planning.
-
Driver-WM: A Driver-Centric Traffic-Conditioned Latent World Model for In-Cabin Dynamics Rollout
Driver-WM is a driver-centric latent world model for causal rollout of in-cabin dynamics conditioned on out-cabin traffic, unifying kinematics forecasting with behavioral and emotional recognition via dual-stream architecture and gated injection.
-
ESARBench: A Benchmark for Agentic UAV Embodied Search and Rescue
ESARBench is the first unified benchmark for MLLM-driven UAV agents that must explore, locate clues, and decide on victim positions in photorealistic simulated SAR environments.
-
Human adaptive variability stabilises collective traffic dynamics
Human car-following variability generates nonlinear damping via non-monotonic time-headway shifts that dissipates traffic perturbations, whereas commercial ACC amplifies them 2.7-5 fold in fuel use.
-
A Distributed Multi-UGV Exploration Framework With Loop-Aware Planning and Descriptor-Aided Localization in Resource-Limited Environments
A distributed multi-UGV framework using descriptor-aided loop closure and loop-aware hierarchical planning reduces exploration time by 15% and travel distance by 14% versus mTSP baseline while achieving 89.9% AR@1 place recognition.
-
DiffSlack: Learning under Nonlinear Inequality Constraints via Learnable Slack Variables
DiffSlack introduces learnable slack variables and a damped Gauss-Newton projection to create a differentiable layer that enforces hard nonlinear inequality constraints in neural network outputs.
-
Hyper-V2X: Hypernetworks for Estimating Epistemic and Aleatoric Uncertainty in Cooperative Bird's-Eye-View Semantic Segmentation
Hyper-V2X uses a Bayesian hypernetwork with partial weight generation and V2X context embedding to produce calibrated epistemic and aleatoric uncertainty estimates for multi-agent BEV segmentation on the OPV2V benchmark.
-
Dr-BA: Separable Optimization for Direct Radar Bundle Adjustment & Localization
Dr-BA delivers a separable optimization approach for direct radar bundle adjustment and cross-session localization using full spinning-radar intensity images, achieving state-of-the-art performance on over 200 km of on-road data.
-
EgoDyn-Bench: Evaluating Ego-Motion Understanding in Vision-Centric Foundation Models for Autonomous Driving
EgoDyn-Bench finds a Perception Bottleneck: foundation models hold ego-motion logic in language but misalign it with vision, underperforming geometric baselines until given explicit trajectories.
-
LiloDriver: A Lifelong Learning Framework for Closed-loop Motion Planning in Long-tail Autonomous Driving Scenarios
LiloDriver uses LLMs and memory-augmented planning in a four-stage pipeline to outperform rule-based and learning-based methods on both common and rare scenarios in the nuPlan benchmark.
-
Comprehensive Robustness Analysis of LiDAR-based 3D Object Detection in Autonomous Driving
Recent LiDAR 3D detectors remain as vulnerable to adversarial attacks as predecessors, with voxel-based and non-anchor-based models showing greater susceptibility under a multi-factor robustness framework.
-
CooperScene: Multi-Modal Cooperative Autonomy Benchmark with C-V2X Communication Characterization
CooperScene provides 59K synchronized frames with 344K 3D annotations from multi-modal sensors on 3 CAVs and 1 RSU plus real C-V2X communication traces for cooperative autonomy benchmarking.
-
Teleoperation Operational Design Domain based on Minimal Risk Maneuver Capability
Proposes basing teleoperation ODD on minimal risk maneuver capability with a dedicated system, demonstrated via use case.
-
Online Adaptive Probabilistic Safety Certificate with Language Guidance
A framework integrates user language and probabilistic environment estimates into adaptive safety certificates that guarantee long-term safety for stochastic systems via probabilistic invariance.
-
Operating Within the Operational Design Domain: Zero-Shot Perception with Vision-Language Models
Vision-language models can serve as zero-shot ODD sensors for autonomous driving when using definition-anchored chain-of-thought prompting with persona decomposition.
-
RadarCNN: Learning-based Indoor Object Classification from IQ Imaging Radar Data
RadarCNN classifies indoor objects from radar IQ data at 97-99% accuracy, holding at ~50% under noise and occlusion.
-
Optimization-based Safe Trajectory Planning for Autonomous Ground Vehicle in Multi-Floor Scenarios
A framework for multi-floor AGV trajectory planning that combines GVD-based task selection with optimization-based trajectory generation and constraint reduction, verified in simulation.
-
Generalization Under Scrutiny: Cross-Domain Detection Progresses, Pitfalls, and Persistent Challenges
A survey that organizes methods for cross-domain object detection into a taxonomy, analyzes domain shift across detection stages, and outlines persistent challenges.