robotics and autonomy

This work addresses the challenge of simultaneously achieving certifiable real-time control and advanced perception capabilities in autonomous flight systems, a balance that existing architectures struggle to maintain due to inherent trade-offs between reliability and flexibility. To bridge this gap, the authors propose a hybrid architecture that integrates NASA’s F´ flight software framework with the ROS 2 middleware, leveraging Protocol Buffers for efficient communication between vision-based navigation and flight control modules. Implemented and validated in closed-loop on an embedded quadrotor platform, this approach represents the first seamless integration of the certifiable F´ framework with the flexible ROS 2 ecosystem. Flight tests totaling 32.25 minutes demonstrate high performance: position estimation at 87.19 Hz, 99.90% data continuity, an average latency of 11.47 ms, only 15.19% CPU utilization, and successful execution of all 15 ground commands, collectively confirming the system’s efficiency, robustness, and certification potential.

Real-time multi-step planning and obstacle avoidance for autonomous robots in dynamic environments remain challenging, particularly under resource constraints and without prior map knowledge. Method: We propose a lightweight, closed-loop reactive planning framework that requires no pre-mapping or offline computation. Our approach integrates biologically inspired attention mechanisms with local LiDAR perception to construct transient control-chain plans. It introduces forward depth-first model checking—novel in real-time multi-step planning—combined with environment-aware 2D LiDAR discretization and closed-loop feedback control. Contribution/Results: The framework provides theoretical guarantees on safety and interpretability. Empirically, it generates safe, multi-step local trajectories within 100 ms on low-power embedded hardware. In complex scenarios—including dead ends and playgrounds—it significantly outperforms single-step reactive systems in obstacle avoidance success rate and response robustness.

To address challenges in high-speed, high-dynamics racing environments—namely, low perception robustness, difficulty in trajectory optimization, and poor simulation-to-reality transfer—this work designs and implements a ROS 2-based autonomous driving software stack for the Formula Student UK Driverless 2025 competition. The method integrates multi-modal sensing (360° LiDAR, ZED2i stereo vision, and GNSS/IMU), incorporates vehicle dynamics modeling, and enforces environmental constraints during trajectory generation. Developed in the CARLA simulator and deployed in real time on a Jetson AGX Orin edge platform, the system achieves stable cone detection within 35 meters and robust closed-loop control. Key contributions include a lightweight perception-planning co-design architecture tailored to racing scenarios, enabling seamless transfer from simulation to physical hardware—including chassis and actuators—and validated via real-vehicle closed-loop testing. This significantly enhances autonomous performance on complex tracks and improves engineering deployability.

To address the stringent real-time and high-precision navigation requirements of autonomous racing on closed-circuit tracks, this paper proposes a modular autonomous driving architecture. The system decouples environment perception, SLAM-based localization and mapping, optimal trajectory generation, and model predictive control (MPC) into independent, interoperable subsystems, coordinated via a unified low-latency data pipeline—enhancing scalability and real-time performance. A key innovation lies in the fusion of multi-source visual cues with high-definition maps to achieve centimeter-level localization and millisecond-scale closed-loop control under constrained computational resources. Experimental validation on complex racetracks demonstrates robust high-speed trajectory tracking and dynamic obstacle avoidance, achieving an average lateral tracking error of <0.15 m and a control frequency of 50 Hz. These results confirm the system’s reliability and engineering practicality for competitive autonomous racing applications.

This work addresses the challenges commonly encountered in multirotor flight control research—namely, the complexity of simulation-to-hardware workflows, poor code readability, and limited extensibility—by presenting a lightweight, modular, full-stack open-source flight control system built on ROS 2 and ROSflight 2.0. The proposed architecture enables seamless deployment between simulation and real hardware while significantly simplifying code structure without compromising performance. Its high modularity enhances both readability and extensibility, facilitating rapid research iteration. Experimental results demonstrate that the system achieves waypoint tracking performance comparable to state-of-the-art flight controllers, yet with a smaller, cleaner codebase that streamlines development and validation in academic settings.

This work addresses the challenge of autonomous operation for heterogeneous robotic platforms—specifically aerial and legged robots—in complex environments where GNSS is denied and perceptual conditions are degraded. The authors propose a unified autonomous system architecture that integrates multimodal sensing (LiDAR, radar, vision, and inertial measurements), factor-graph-based SLAM, semantic understanding, scale-adaptive motion planning, and a multi-layer safety mechanism grounded in control barrier functions. For the first time, this framework enables end-to-end coordination among perception, planning, and safety within a single pipeline. Experimental validation on rotorcraft drones and legged robots demonstrates robust performance in navigating, exploring, detecting targets, and conducting inspections under challenging conditions such as smoke and self-similar structures. The associated code and dataset have been publicly released.

This work addresses the challenge of coordinating global path tracking and local obstacle avoidance under noisy, delayed, or missing LiDAR perception to ensure the safety and robustness of autonomous driving systems. The authors propose a ROS 2-native continuous gating fusion mechanism that dynamically weights Ackermann steering commands from a Pure Pursuit global controller and a LiDAR Gap Follow local controller using a PPO-trained policy network operating on compact perceptual features, complemented by a safety verification module. This approach enables interpretable behavior coordination without modifying the underlying controllers and introduces a custom LiDAR error injection protocol for systematic robustness evaluation. Experiments demonstrate that, in close-proximity overtaking scenarios, the method significantly improves safety success rates over a lightweight predictive baseline as perception errors intensify, while maintaining real-time control performance.

This work addresses the validation gap between simulation and real-world deployment of autonomous driving algorithms, particularly the lack of efficient, high-fidelity testing platforms for safety-critical scenarios. To bridge this gap, the authors propose a mixed-reality hardware-in-the-loop testing framework that seamlessly integrates physical mobile robots with high-fidelity virtual environments, enabling multimodal sensing, vehicle-to-everything (V2X) communication, and large-scale multi-agent collaboration. A key innovation is the coexistence of physical and virtual agents within a unified architecture, coupled with an online learning controller based on control barrier functions (CBFs) that establishes an integrated perception-planning-control safety assurance mechanism. Experimental results demonstrate that the platform significantly enhances the reliability and efficiency of sim-to-real transfer and validates its effectiveness across diverse safety-critical scenarios.

This work addresses the limited generalizability and transferability of current foundation models in robotics, which often require custom integration of perception, actuation, and safety mechanisms. To overcome this, the authors propose ROSClaw—a model-agnostic execution layer that enables plug-and-play deployment of arbitrary foundation models on any ROS 2 robot by integrating the OpenClaw agent runtime with ROS 2. Key innovations include standardized capability discovery, multimodal observation normalization, action validation within configurable safety bounds, and structured audit logging. Experiments across three robotic platforms and four foundation models demonstrate up to a 4.8× difference in non-policy action proposal rates and show that the proposed execution layer significantly improves task success rates and safety across diverse frameworks.

This work addresses the challenge of real-time safe autonomous navigation in spatially constrained and dynamically changing environments by proposing a real-time control architecture integrated with 3D LiDAR perception. The approach introduces an ellipsoidal safety region aligned with the robot’s body geometry, which rotates with the robot’s pose in the world frame to generate time-varying obstacle avoidance constraints. A dedicated time-varying Control Barrier Function (CBF) is designed for each LiDAR point, enabling efficient handling of numerous constraints at control frequency while minimally interfering with the primary navigation task. Extensive field experiments on a quadrupedal robot demonstrate robust performance in complex scenarios such as narrow underground corridors, where the system reliably copes with dynamic obstacles, unreliable high-level commands, and abrupt localization shifts, thereby validating its high reliability and practicality.