End-to-End (E2E) autonomous driving models have shown growing capability in recent years, with performance improving on increasingly challenging benchmarks. However, modern generative E2E planners still suffer from a substantial number of catastrophic failures in safety-critical scenarios. We find that many such failures arise from violations of physical constraints and safety requirements, leading to unsafe behavior. Motivated by this finding, in this paper, we focus on improving safety outcomes in generative end-to-end driving with a targeted reduction of catastrophic planning failures, instead of enhancing average planning quality. Towards this end, we propose DriveSafer, a failure-aware safety framework for end-to-end planners. DriveSafer explicitly steers generative planners towards safe behaviors leveraging both training-time safety constraints and inference-time safety guidance. Compared to the state-of-the-art DiffusionDrive model, on the NAVSIM benchmark, DriveSafer reduces the number of catastrophic failures (PDMS=0) by 48%, with over 65% reduction in drivable-area compliance failures.
As the market for commercially available unmanned aerial vehicles (UAVs) booms, there is an increasing number of small, teleoperated or autonomous aircraft found in protected or sensitive airspace. Existing solutions for removal of these aircraft are either military-grade and too disruptive for domestic use, or compose of cumbersomely teleoperated counter-UAV vehicles that have proven ineffective in high-profile domestic cases. In this work, we examine the use of a quadrotor for autonomously targeting semi-stationary and moving aerial objects with little or no prior knowledge of the target's f…
METDrive: Multi-modal End-to-end Autonomous Driving with Temporal Guidance
Intelligent Space Robotics Laboratory, Center for Digital Engineering, Skolkovo Institute of Science and Technology（斯克尔科沃科学与技术研究所智能空间机器人实验室，数字工程中心）；Institute of Automation, Qilu University of Technology (Shandong Academy of Sciences)（齐鲁科技大学自动化研究所，山东科学院）
LD-Scene: LLM-Guided Diffusion for Controllable Generation of Adversarial Safety-Critical Driving Scenarios
The Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））；School of Computer Science and Engineering, Sun Yat-sen University（中山大学计算机科学与工程学院）
Technische Universität Braunschweig（布伦瑞克工业大学）；Volkswagen AG（大众汽车集团）
Automated driving has become a major topic of interest not only in the active research community but also in mainstream media reports. Visual perception of such intelligent vehicles has experienced large progress in the last decade thanks to advances in deep learning techniques but some challenges still remain. One such challenge is the detection of corner cases. They are unexpected and unknown situations that occur while driving. Conventional visual perception methods are often not able to detect them because corner cases have not been witnessed during training. Hence, their detection is high…
Arizona State University（亚利桑那州立大学）；Iowa State University（爱荷华州立大学）；George Washington University（乔治华盛顿大学）；University of Michigan（密歇根大学）
Obstacle avoidance for small unmanned aircraft is vital for the safety of future urban air mobility (UAM) and Unmanned Aircraft System (UAS) Traffic Management (UTM). There are many techniques for real-time robust drone guidance, but many of them solve in discretized airspace and control, which would require an additional path smoothing step to provide flexible commands for UAS. To provide a safe and efficient computational guidance of operations for unmanned aircraft, we explore the use of a deep reinforcement learning algorithm based on Proximal Policy Optimization (PPO) to guide autonomous …
Autonomous Systems Lab ETH Zürich（苏黎世联邦理工学院自主系统实验室）
L1 guidance logic is one of the most widely used path following controllers for small fixed-wing unmanned aerial vehicles (UAVs), primarily due to its simplicity (low-cost implementation on embedded on-board processors, e.g. micro-controllers) and ability to track both circles and lines, which make up the vast majority of a typical fixed-wing vehicle's flight plan. The logic was later extended for speed independent dynamic similarity with an adaptive L1 distance, a formulation currently used on common open-source autopilot platforms. Two primary drawbacks remain, specific to small, slow flying…
Loiter UAV Reinsertion Guidance for Fixed-wing UAV Corridors
American Institute of Aeronautics and Astronautics, Inc.（美国航空航天学会）；Department of Mechatronics Engineering（机电工程系）；Department of Aerospace Engineering（航空航天工程系）
Navigation of mobile robots within crowded environments is an essential task in various use cases, such as delivery, health care, or logistics. Deep Reinforcement Learning (DRL) emerged as an alternative method to replace overly conservative approaches and promises more efficient and flexible navigation. However, Deep Reinforcement Learning is limited to local navigation due to its myopic nature. Previous research works proposed various ways to combine Deep Reinforcement Learning with conventional methods but a common problem is the complexity of highly dynamic environments due to the unpredic…
High-Speed Vision-Based Flight in Clutter with Safety-Shielded Reinforcement Learning
Institute of Cyber-Systems and Control, College of Control Science and Engineering, Zhejiang University（浙江大学控制科学与工程学院工业控制技术研究所）；Differential Robotics（微分机器人）
Diffusion Forcing Planner: History-Annealed Planning with Time-Dependent Guidance for Autonomous Driving
University of Science and Technology of China（中国科学技术大学）；Yinwang Intelligent Technology Co., Ltd（银网智能科技有限公司）；Hefei Institutes of Physical Science, Chinese Academy of Sciences（中国科学院合肥物质科学研究院）
Department of Mechanical and Civil Engineering, Caltech（加州理工学院机械与土木工程系）；Department of Computing and Mathematical Sciences, Caltech（加州理工学院计算与数学科学系）；Department of Mechanical Engineering, Tufts University（塔夫茨大学机械工程系）
Robotic systems navigating in real-world settings require a semantic understanding of their environment to properly determine safe actions. This work aims to develop the mathematical underpinnings of such a representation -- specifically, the goal is to develop safety filters that are risk-aware. To this end, we take a two step approach: encoding an understanding of the environment via Poisson's equation, and associated risk via Laplace guidance fields. That is, we first solve a Dirichlet problem for Poisson's equation to generate a safety function that encodes system safety as its 0-superleve…
Safety-Critical Input-Constrained Nonlinear Intercept Guidance in Multiple Engagement Zones
Unmanned Systems Lab, Department of Electrical and Computer Engineering, The University of Texas at San Antonio（无人系统实验室，电气与计算机工程系，德克萨斯大学圣安东尼奥分校）；Guidance, Autonomy, Learning, and Control for Intelligent Systems (GALACxIS) Lab, Department of Aerospace Engineering and Engineering Mechanics, University of Cincinnati（智能系统引导、自主、学习和控制（GALACxIS）实验室，航空航天工程与工程力学系，辛辛那提大学）
CNRS IRL 2010 CROSSING（法国国家科学研究中心国际联合研究实验室CROSSING）；ENSTA Bretagne（法国国立高等先进技术布列塔尼分校）；Flinders University（弗林德斯大学）；University of South Australia（南澳大利亚大学）；University of Adelaide（阿德莱德大学）
Karlsruhe Institute of Technology（卡尔斯鲁厄理工学院）；Karlsruhe University of Applied Sciences（卡尔斯鲁厄应用科学大学）
Hand guidance emerged from the safety requirements for collaborative robots, namely possessing joint-torque sensors. Since then it has proven to be a powerful tool for easy trajectory programming, allowing lay-users to reprogram robots intuitively. Going beyond, a robot can learn tasks by user demonstrations through kinesthetic teaching, enabling robots to generalise tasks and further reducing the need for reprogramming. However, hand guidance is still mostly relegated to collaborative robots. Here we propose a method that doesn't require any sensors on the robot or in the robot cell, by using…
Autonomous Vehicles (AVs) are complex systems that drive in uncertain environments and potentially navigate unforeseeable situations. Safety of these systems requires not only an absence of malfunctions but also high performance of functions in many different scenarios. The ISO/PAS 21448 [1] guidance recommends a process to ensure the Safety of the Intended Functionality (SOTIF) for road vehicles. This process starts with a functional specification that fully describes the intended functionality and further includes the verification and validation that the AV meets this specification. For the …
