https://hu-zlatan.github.io/embodied-ai-radar/robots/aerial.xml 2026-10-29T00:00:00ZEmbodied AI Radar https://arxiv.org/abs/2610.37557 2026-10-28T00:00:00Z

Threading a quadrotor through a moving narrow gap is a demanding test of agile flight, requiring centimeter-level spatial alignment and precise temporal execution. We address this challenge using a control policy that operates directly on onboard visual and inertial measurements to output collective thrust and body-rate commands. The task is difficult because success lies on a thin feasible manifold: small errors in timing, position, attitude, or visual observability cause collision. Prior planning and learning methods rely on privileged state, teachers, planners, or distillation; learning …

https://arxiv.org/abs/2610.37543 2026-10-28T00:00:00Z

Event cameras offer low latency, high dynamic range, and milliwatt-level power consumption, making them ideal for agile robots such as tiny drones navigating through cluttered environments. However, existing simulators generally render photorealistic images before converting them into events, making large-scale reinforcement learning prohibitively slow. We propose an analytic event simulator that exploits what event cameras fundamentally measure - moving contrast edges - to synthesize event streams directly from scene geometry and ego-motion without visual rendering. Our simulator generates a …

https://arxiv.org/abs/2610.37470 2026-10-28T00:00:00Z

Medium altitude, long endurance unmanned aerial systems such as Shahed-type strike drones are an unanswered threat in modern conflicts. Conventional air defense systems are too expensive to provide a feasible large-scale solution. Quadcopter interceptors are a cheap, battle-proven alternative, but their extremely limited flight time restricts them to low altitude targets at short range, whereas fixed-wing platforms have the endurance to reach higher altitude targets from longer distances. However, such interceptors are still operated manually. Recent developments in learning-based controllers …

https://arxiv.org/abs/2610.37190 2026-10-28T00:00:00Z

Flapping-wing micro air vehicles are prized for quiet flight, yet the few existing measurements show engineered wings to be 15-30 dB louder than the animals they imitate. This paper investigates two passive geometric approaches to reduce the acoustic signature of a flapping membrane wing, and applies these to an exemplary wing. A significant sound source in flapping wings is the impulsive click radiated at stroke reversal, when a slack membrane snaps between two curvatures. Our first noise-reduction approach optimises the wing's venation layout, based on an analytical model of the reversal …

https://arxiv.org/abs/2610.37111 2026-10-28T00:00:00Z

Active simultaneous localization and mapping (SLAM) addresses the problem of planning robot motion to build complete and accurate maps of unknown environments through efficient exploration while maintaining high localization accuracy. This raises a dilemma between exploring unknown spaces to expand coverage and exploiting visited regions to perform active loop closures. Despite numerous prior works on ground robots, aerial robots with high agility and constrained sensor suites face more significant localization drift and thereby demand dedicated active SLAM planning. Moreover, a unified …

https://arxiv.org/abs/2610.36968 2026-10-28T00:00:00Z

Image-goal navigation (ImageNav) requires an agent to reach a location specified by a goal image in an unknown environment. For drones, this task demands exploration, goal- directed navigation, and collision avoidance under limited onboard computing resources. Existing aerial ImageNav methods are built on single-view observations, requiring additional rotation and exploration to search for goal-relevant cues outside the field of view. In this paper, we present Omni-IGN, a fully onboard ImageNav system that integrates quad-fisheye perception, an attention-based reinforcement learning (RL) …

https://arxiv.org/abs/2610.36450 2026-10-28T00:00:00Z

Drone-based 3D reconstruction and inspection require viewpoints that satisfy image-quality constraints while observing each surface region multiple times, which overlap-based planning heuristics do not guarantee. We study ground sample distance (GSD)-constrained $k$-coverage viewpoint selection for structures represented by surface voxels. Candidate viewpoints are generated along outward surface directions, screened with distance, viewing-angle, field-of-view, and occlusion tests, and augmented for voxels with insufficient visibility. Selection is then posed as set multi-cover with per-voxel …

https://arxiv.org/abs/2609.15447 2026-10-28T00:00:00Z

Combining aerial thrust with spring-loaded hopping makes monopedal quadcopters promising for locomotion over complex terrain, but heuristic proportional-integral-derivative (PID) tuning limits coordination between active thrust and passive contact dynamics. We present a direct estimated-state-to-motor Proximal Policy Optimization (PPO) policy that commands four motors without an explicit hopping state machine or low-level attitude PID. Its reward combines Energy-Manifold Shaping for mass-normalized vertical-energy tracking and apex-state anchoring with Efficiency Shaping, which uses a …

https://arxiv.org/abs/2610.35342 2026-10-27T00:00:00Z

While unmanned aerial vehicles (UAVs) are increasingly deployed for pipeline inspection, prevailing research predominantly assumes static pipes. In real-world industrial environments, pipes often exhibit dynamic movements, such as vibrations. This imposes more stringent requirements on UAVs' adaptive control and computational efficiency than in static scenarios. Therefore, this paper proposes a servo control framework named VisGeo-PipeTrack, which utilizes geometric constraints to enable UAV tracking of dynamic pipes. To achieve efficient UAV positioning, YOLOv5 is employed to detect labels …

https://arxiv.org/abs/2610.35165 2026-10-27T00:00:00Z

Quadrotor controller comparisons commonly report a single tracking error, although the resulting ranking also depends on tuning effort, reference information, actuator prediction, optimizer fallback, and command timing. We quantify these effects for cascade PID, hover LQR, boundary-layer sliding-mode control (BL-SMC), linear MPC with LQR fallback (LMPC-F), a preview-enabled LMPC-F, and a geometric $SO(3)$ controller on a common nonlinear plant with rotor gyroscopic coupling, motor lag, and saturation. A primary study (556 episodes at 50~Hz) combines matched twelve-candidate tuning, 400 paired …

https://arxiv.org/abs/2610.34873 2026-10-27T00:00:00Z

This paper studies UAV swarm search in GNSS-denied maritime environments, where inter-UAV ranging drives cooperative localization and uncertainty propagates along the multi-hop reference chain, affecting connectivity, collision avoidance, and search effectiveness. We propose an uncertainty-aware control barrier function (CBF) framework approximating cascading localization covariance through Fisher information matrices. First, the cascading covariance model captures how ranging noise and neighbor uncertainty accumulate along the localization chain. Second, uncertainty-tightened CBFs …

https://arxiv.org/abs/2610.34789 2026-10-27T00:00:00Z

Quadrotor interceptors equipped with strapdown cameras offer a low-cost solution for physically countering unauthorized UAVs. However, this configuration inherently couples vehicle maneuvering and camera pointing, creating a conflict between rapid closing and maintaining target visibility. We present interception-oriented spherical-bearing nonlinear model predictive control (ISB-NMPC), which couples spherical-bearing prediction with nonlinear quadrotor dynamics to jointly optimize target-directed closing motion and visibility. An attitude-compensated extended Kalman filter on S^2 estimates …

https://arxiv.org/abs/2610.34538 2026-10-27T00:00:00Z

Quadrotor trajectory tracking under unmodeled disturbances often relies on online residual adaptation, but short out-domain residuals can corrupt the learned disturbance model when constant-step updates treat every residual gradient uniformly. We propose Dual Gradient Consensus (DGC), an adaptive-learning method for mixed in-domain/out-domain disturbances that retains the random Fourier feature model, offline BPTT policy, and least-squares adaptation of a prior domain-adaptive residual policy (DAPL) while modifying the online update step size. DGC compares each residual gradient with long- …

https://arxiv.org/abs/2610.34456 2026-10-27T00:00:00Z

In this project, I explored controller design and trajectory generation for a quadrotor. Differential flatness was used for trajectory generation based on one of the references. A minimum snap trajectory is generated from the given algorithm based on waypoint navigation, and a simple proportional plus derivative (PD) controller is used to track the trajectories. Simulation results are presented which generate smooth trajectories for position and velocity of the quadrotor.

https://arxiv.org/abs/2610.34454 2026-10-27T00:00:00Z

This paper presents two controllers for a quadrotor equipped with a two-DOF robot manipulator that physically interacts with the environment. Cartesian impedance and adaptive sliding mode controllers are implemented and compared for a simple hovering task in the presence of some bounded disturbance. A full dynamic model is developed, and the two above controllers are implemented with the same initial conditions and desired position.

https://arxiv.org/abs/2610.34054 2026-10-27T00:00:00Z

This paper proposes a command-aligned sampling distribution that incorporates navigation commands directly into the exploration process to improve the efficiency of sampling-based motion planning for quadcopters in three-dimensional environments. Inspired by the Language-Aided Subset Sampling Motion Planner, this paper extends command-aware subset sampling to aerial systems. Subset sampling is generalized to command-aligned regions, enabling structured 3D exploration while preserving probabilistic completeness. Natural language instructions are grounded into yaw-conditioned and altitude-aware …

https://arxiv.org/abs/2610.33906 2026-10-27T00:00:00Z

When a team of drones explores a building together, each drone tells the others which frontier it is going to and shares the map it has built. This avoids duplicated effort, but it also means that the team trusts every identity from which it receives messages. In a Sybil attack, one compromised drone creates fake identities, and these fake teammates claim frontiers and send fabricated scans, so the real drones leave parts of the building unexplored and end up with walls in their maps that do not exist. We study five such attackers, including one that seals off the openings between floors, and …

https://arxiv.org/abs/2610.33640 2026-10-27T00:00:00Z

Continuous aerial tracking often requires multiple drones to maintain target coverage over long periods. During handoff, the receiving UAV may lack prior target observations and must actively acquire a usable view without disrupting the sender's ongoing coverage. This work introduces HARP, a handoff-aware recovery planner that registers bounded sender predictions in the receiver's local map, distinguishes field-of-view failure from physical occlusion, and generates failure-specific yaw, translation, and parallax viewpoints. Candidate motions are checked for receiver clearance, sender-fiducial …

https://arxiv.org/abs/2610.32645 2026-10-26T00:00:00Z

Indoor UAV localization with minimal onboard sensing remains challenging, particularly in dynamic environments. We propose augmented filtered Monte Carlo localization (AF-MCL) for 2D localization using only four single-point range sensors and measured heading. AF-MCL rejects range measurements inconsistent with the known static map and adaptively injects particles to recover from localization divergence and kidnapping. Beam classification is performed only at the previous position estimate, making its ray-casting cost independent of the particle count. The particle distribution additionally …

https://arxiv.org/abs/2610.32559 2026-10-26T00:00:00Z

Supervising a multi-UAV reconnaissance mission usually means plotting waypoints and watching every drone. We present UAV SwarmAgent, a hierarchical multi-agent system in which an operator states the mission in one natural-language sentence and a ground-station agent writes and executes code that turns it into per-UAV routes. The agent grounds inspection targets in an aerial map with a vision-language model, splits them among N UAVs by balancing route length, and dispatches worker agents that verify incidents with a compact onboard decision model. In simulation over 4 incident types and 3 …

https://arxiv.org/abs/2610.31617 2026-10-23T00:00:00Z

Tasks such as quadrotor dynamic-gate racing require agents to anticipate external motion while maintaining precise high-frequency control. World-model-based reinforcement learning provides a natural solution by learning latent dynamics for future imagination, yet it often entangles environmental evolution and the robot's own dynamics in a single learned rollout, where model bias can affect policy optimization efficiency. We propose DiffDreamer, a differentiable reinforcement learning framework that tightly couples a learned world model with differentiable simulation for actor optimization. In …

https://arxiv.org/abs/2610.31051 2026-10-23T00:00:00Z

Tilt-rotor quadcopters redirect thrust for translation without banking, but complicate modeling, state estimation, and disturbance rejection. This paper presents a simulation-based flight stack for a \SI{2.00}{\kilo\gram} tilt-rotor quadcopter. The system uses a six-degree-of-freedom nonlinear model with first-order motor dynamics and two-axis rotor tilt, parameterized using a 3D CAD model. A cascaded PD controller commands forward and lateral tilt angles. A 14-state EKF fuses sensor data across hover, early-transition, and transition modes using a hysteretic (Schmitt-trigger) classifier to …

https://arxiv.org/abs/2610.28950 2026-10-22T00:00:00Z

We present HyperDrone, an extremely fast quadrotor simulator focused on high-performance batch rendering that is more than 20x faster than existing state-of-the-art batch quadrotor simulators. HyperDrone combines highly optimized multirotor dynamics with HyperLux, a new ray-tracing engine that supports Nvidia OptiX, Vulkan, Apple Metal and WebGPU backends that takes full advantage of modern ray-tracing hardware, capable of running both components entirely on the GPU. HyperLux renders RGB, depth, semantic/instance segmentation, surface normals and optical flow observations at configurable …

https://arxiv.org/abs/2610.27414 2026-10-21T00:00:00Z

Reliable quadrotor flight in disturbed environments requires predictive models that remain accurate as unmodeled aerodynamic effects change during deployment. We introduce a disturbance-aware Koopman Control Family (DA-KCF) framework that preserves an offline-identified nominal model while learning only a disturbance-induced correction online. A reference-dependent KCF is first learned from simulation data and calibrated using limited hardware data. During deployment, the nominal KCF remains fixed while a lightweight residual ensemble learns a low-dimensional disturbance correction from …

https://arxiv.org/abs/2610.27315 2026-10-21T00:00:00Z

The integration of dynamic legged locomotion and aerial mobility offers versatility that is deterred by challenges in payload distribution, actuation, and the contrasting requirements of both domains. In this paper, we present the design and operation of a multimodal quadrupedal aerial-legged robot. The presence of legged aerial systems also allows for thruster-assisted legged locomotion, which we make possible with a unified control framework using Control Contraction Metrics (CCM). We search for separate domain-specific contraction metrics for walking and flight independently offline and …

https://arxiv.org/abs/2610.26810 2026-10-20T00:00:00Z

This paper presents the hardware implementation of a quadratic program-based controller on Harpy: thruster-assisted bipedal robot to perform thruster-assisted trotting. This is inspired by wing-assisted locomotion observed in birds such as the chukar partridge, which employ thrust vectoring to manipulate the ground reaction force. The proposed controller formulates the thruster-assisted spring-loaded inverted pendulum dynamics as a condensed QP-based model predictive control problem, enabling simultaneous regulation of ground contact force and active thrust vectoring at 500 Hz. A decoupled …

https://arxiv.org/abs/2610.26508 2026-10-20T00:00:00Z

Existing high-speed navigation systems for micro aerial vehicles (MAVs) commonly rely on depth cameras, stereo vision, or LiDAR to obtain explicit geometric information, while monocular cameras offer a lighter and lower-cost alternative. However, monocular flight remains challenging because RGB images offer no direct metric cues and visual appearance varies considerably across environments. We propose a geometry-aware monocular visual planning framework that predicts trajectories from two image frames, their relative pose, and the MAV state. Without explicitly reconstructing depth, the …

https://arxiv.org/abs/2610.25869 2026-10-20T00:00:00Z

Reliable recognition of mission-level UAV intent from multimodal observations is essential for low-altitude airspace surveillance and security. Existing multimodal UAV datasets primarily support detection, tracking, and localization, leaving limited support for evaluating mission-intent recognition across held-out trajectory programs and incomplete radar histories. We introduce AeroIntent, a simulation-based benchmark comprising 990 synchronized multimodal sequences generated by 22 trajectory programs across five mission intents. The benchmark provides radar, RGB, infrared, and depth …

https://arxiv.org/abs/2610.25688 2026-10-20T00:00:00Z

Direct wind sensing from a multi-rotor UAV enables localized atmospheric measurements, but the reliability of the anemometer depends on vehicle motion. This work presents a vehicle-state-aware confidence method that uses horizontal and vertical vehicle speed to quantify the acquisition of each wind observation. The resulting confidence score is used to adapt measurement influence to a two-dimensional vector estimator. Experiments were conducted during horizontal translation, vertical motion, and held-out autonomous waypoint flight. The proposed two-feature formulation achieved an ROC-AUC of …

https://arxiv.org/abs/2610.25455 2026-10-20T00:00:00Z

Classical robot control, Finite State Machines and Behavior Trees (BTs), is deterministic and inspectable but fixed at design time: it fails on situations its designers did not anticipate. Foundation models can synthesize new control logic at runtime, at a cost in latency and non-determinism hard to accept in safety-critical aerial systems. Existing LLM-driven approaches keep the model in the loop permanently and discard what it produces. We propose Self-Crystallizing Behavior Trees (SCBT), in which the foundation model is a compiler rather than a runtime. When a BT node fails, an LLM …

https://arxiv.org/abs/2610.25407 2026-10-20T00:00:00Z

Onboard inference latency creates a mismatch between the trajectories predicted by vision-language-action (VLA) policies and the ongoing motion of unmanned aerial vehicles (UAVs). We present AeroDASH, an asynchronous VLA framework for continuous language-conditioned flight. A B-spline policy is initialized through supervised fine-tuning on expert demonstrations, providing continuous trajectory references in position and yaw. During execution, trajectory fusion removes the expired prediction prefix and connects the remaining segment to the active reference with derivative continuity while …

https://arxiv.org/abs/2610.25024 2026-10-20T00:00:00Z

Multi-quadrotor passage planning requires coordinated spatial routes and passage times under vehicle dynamics and actuation constraints. Optimizing these decisions jointly couples temporal allocation, nonlinear motion, and inter-vehicle interactions, creating a computational challenge for local trajectory generation. We propose a learning-optimization framework that predicts spatial waypoints and ordered passage events for independently solved quadrotor planners. A common physical time grid aligns the vehicles, with intermediate events fixed by the networks and terminal progress optimized …

https://arxiv.org/abs/2610.23878 2026-10-20T00:00:00Z

Aerial tracking of autonomous ground vehicles can provide useful situational awareness in environments where fixed external perception infrastructure is unavailable. However, visual-only tracking from an unmanned aerial vehicle (UAV) is limited by camera field of view, viewpoint changes, occlusions, and temporary loss of target visibility. This paper presents a heterogeneous UAV--UGV methodology for aerial tracking of small-scale autonomous ground vehicles in simulation. The system is implemented in ROS~2 and Gazebo Sim using a Tello-like UAV equipped with a front-facing RGB camera and …

https://arxiv.org/abs/2610.23877 2026-10-20T00:00:00Z

Integrating Vision-Language Models (VLMs) into autonomous UAV navigation enables high-level semantic reasoning but introduces critical temporal bottlenecks that can destabilize flight control. This paper presents a systematic benchmark that evaluates the operational trade-offs among model scale, hardware constraints, and flight reliability. By subjecting the system to extensive stress tests on both a high-performance workstation and a constrained integrated-graphics laptop, we isolate the kinetic impact of inference latency against reasoning accuracy. Our findings reveal a performance …

https://arxiv.org/abs/2610.23635 2026-10-19T00:00:00Z

Obstacle avoidance for autonomous quadrotors in cluttered environments remains challenging, especially when both dense static obstacles and moving obstacles are present. Existing methods are often tailored to either static or dynamic settings, which always limit their adaptability across diverse environments. This paper presents TransNav, a transformer-based end-to-end navigation method for quadrotor obstacle avoidance in both static and dynamic environments. TransNav employs a patch-based point-cloud encoding scheme that combines local feature extraction with global contextual modeling, …

https://arxiv.org/abs/2610.22225 2026-10-15T00:00:00Z

Propulsive drones are a promising new technology for generating high-resolution terrain models of the Moon. In future lunar missions, propulsive drones will complete a series of hops across the lunar surface while capturing images with downward-facing cameras. We present an autonomous motion planning framework that uses a vision-guided optimal control strategy for performing safe, efficient hops between specified waypoints. The drone plans parabolic hops between waypoints generated by a global-level planner; if a nominal waypoint is unsafe or unreachable, the drone then completes a diversion …

https://arxiv.org/abs/2610.22092 2026-10-15T00:00:00Z

Aerial robots operating near people must treat a pedestrian, a parked car, and a wall differently, yet conventional perception--planning stacks reduce all of them to occupied geometry---a \emph{human-aware safety gap} in which category-specific clearance is unavailable to the planner and temporary occlusions remove pedestrian-specific constraints from its map. We present a modular, fully onboard pipeline that closes this gap by turning visual observations into context-dependent safety constraints for real-time UAV navigation. A shared BLIP encoder drives scene captioning and text-conditioned …

https://arxiv.org/abs/2610.22057 2026-10-15T00:00:00Z

Actively tilting aerial robots can generate six-dimensional wrenches, but slow servo response can make a desired wrench infeasible during reconfiguration. This paper develops a control barrier function (CBF) allocator that preserves wrench feasibility under bounded servo-configuration mismatch. An exact equivalent-wrench identity connects the thrust required at nominal and actual configurations. For a nonsingular six-channel allocation map, inverse allocation reduces robust wrench containment to twelve thrust margins tightened by an ellipsoidal reserve. Their complete derivatives yield a …

https://arxiv.org/abs/2610.21978 2026-10-15T00:00:00Z

Biological visual navigation achieves remarkable robustness at extremely low power, whereas frame-based robotic systems relying on artificial neural networks remain limited in both respects, particularly for energy-constrained micro aerial vehicles (MAVs) operating in complex environments. Inspired by the event-driven, sparse, and asynchronous principles of biological visual systems, neuromorphic sensing and computation provide a promising alternative for robust and energy-efficient navigation. In this paper, we propose the first end-to-end event-driven neuromorphic paradigm for MAV visual …

https://arxiv.org/abs/2610.20823 2026-10-14T00:00:00Z

A mesh of interceptors defending a fixed site must spread over the predicted arrival point of an approaching object. This is a recurring task in multi-agent robotics: N agents, each covering a disc of radius r, spread over a probability density so that a point drawn from it falls within some agent's disc. We give closed forms for the quantities that bound performance before any policy is chosen, and validate each by Monte Carlo. The uncertainty of a point predicted by extrapolating a noisy linear track follows from least-squares variance propagation plus a delta-method term for the uncertain …

https://arxiv.org/abs/2610.18850 2026-10-13T00:00:00Z

Collaborative planning under operational constraints is essential for heterogeneous robot teams in large-scale missions. Unmanned Aerial Vehicles (UAVs) provide rapid area coverage but are limited by onboard energy, while Unmanned Ground Vehicles (UGVs) can serve as mobile recharging stations yet are constrained by road-network traversability. Effective collaboration must enforce operational constraints during execution while enabling parallel, coordinated actions. Masked Diffusion Models (MDMs) have shown promise through token unmasking but have two key limitations: entropy-guided scheduling …

https://arxiv.org/abs/2610.18425 2026-10-13T00:00:00Z

Cooperative aerial manipulation is difficult under wind, payload motion, and noisy goal broadcasts because multiple Unmanned Aerial Vehicles (UAVs) are coupled through a shared payload. We present CALM (Cooperative Aerial Learned Manipulation), a decentralized controller for aerial manipulators carrying a basket while tracking a moving ground robot. CALM combines a parameter-shared Multi-Agent Reinforcement Learning (MARL) policy, trained using Multi-Agent Proximal Policy Optimization (MAPPO), with goal estimation, per-UAV Extended State Observers (ESOs), basket-mass estimation, and …

https://arxiv.org/abs/2610.18381 2026-10-13T00:00:00Z

Aerial mobile manipulation requires unmanned aerial vehicles (UAVs) to traverse heterogeneous environments such as indoor and outdoor domains, and connect navigation with subsequent physical interaction within the same trajectories. Existing aerial benchmarks are typically confined to single-environment settings, limiting evaluation of agents across continuous environmental transitions. When interaction is evaluated, agents are often initialized near the target, leaving the transition from navigation to interaction largely unassessed. This work introduces CrossFly, a benchmark for …

https://arxiv.org/abs/2610.17419 2026-10-12T00:00:00Z

Autonomous quadrotor flight in cluttered environment is an key capability for tasks like inspection and search. Navigation policies trained in simulation and transferred to hardware can fly at high speed in open space, but they can still collide as obstacle clearance shrinks. We attribute most of these collisions to a single mechanism, namely motion in directions the onboard camera does not observe. Heading that is not aligned with the direction of travel and lateral excursion during obstacle avoidance both carry the vehicle toward obstacles outside its field of view. We introduce ClearNav, a …

https://arxiv.org/abs/2610.16532 2026-10-12T00:00:00Z

This work develops a geometric theory of aerodynamic promptness for multirotor platforms whose propeller directions depend on a smooth orientation-mechanism configuration. The actuator state consists of the actual signed propeller speeds and the mechanism configuration, and the nonlinear actuator map retains both the signed-quadratic propeller law and the configuration-dependent thrust directions. Constant direction maps represent fixed propellers, while shared dependence on mechanism coordinates represents mechanically coupled tilts. A Riemannian metric on the product actuator manifold …

https://arxiv.org/abs/2610.16119 2026-10-12T00:00:00Z

Differentiable simulation enables efficient policy optimization through analytic gradients, yet safety is commonly incorporated as a weighted penalty, requiring manual tuning of the trade-off between task performance and constraint satisfaction. We instead formulate obstacle avoidance as an explicit constraint and train quadrotor policies using primal--dual gradient descent--ascent through a differentiable simulator. We introduce two key components to enable effective constrained training. First, we propose a bounded clearance-based constraint function that provides informative gradients …