Typical LiDAR SLAM architectures feature a front-end for odometry estimation and a back-end for refining and optimizing the trajectory and map, commonly through loop closures. However, loop closure detection in large-scale missions presents significant computational challenges due to the need to identify, verify, and process numerous candidate pairs for pose graph optimization. Keyframe sampling bridges the front-end and back-end by selecting frames for storing and processing during global optimization. This article proposes an online keyframe sampling approach that constructs the pose graph using the most impactful keyframes for loop closure. We introduce the Minimal Subset Approach (MSA), which optimizes two key objectives: redundancy minimization and information preservation, implemented within a sliding window framework. By operating in the feature space rather than 3-D space, MSA efficiently reduces redundant keyframes while retaining essential information. Evaluations on diverse public datasets show that the proposed approach outperforms naive methods in reducing false positive rates in place recognition, while delivering superior ATE and RPE in metric localization, without the need for manual parameter tuning. Additionally, MSA demonstrates efficiency and scalability by reducing memory usage and computational overhead during loop closure detection and pose graph optimization.

In this paper, we propose a holistic method for coordinated task assignment and trajectory generation for multi-agent systems in a known environment. First, we propose CBS-TS that generates conflict-free paths for each robot, while ensuring all the goal locations are visited. CBS-TS conducts the best-first search over a conflict forest, which is guaranteed to be optimal and complete. Then, the resulting paths are optimized to dynamically feasible trajectories. Through optimization constraints, the smoothed trajectory remains conflict-free and ensures all goal locations are precisely visited. We provide extensive simulation results to analyze the computational efficiency of the proposed algorithm and conduct physical experiments with heterogeneous robots to demonstrate the feasibility of generated trajectories.

This article presents an experimental demonstration of an attitude-constrained adaptive guidance and control framework for safe autonomous docking, evaluated using a planar floating satellite platform testbed. The proposed approach combines a jerk-minimizing explicit guidance law, which enforces terminal constraints on position, velocity, and acceleration, with an adaptive tracking controller designed to handle modeling uncertainties, actuator limitations, and external disturbances. The guidance strategy generates a smooth, real-time trajectory for the chaser satellite, ensuring compatibility with limited onboard computation and maintaining high terminal accuracy. To ensure safe operation throughout the docking maneuver, a barrier-Lyapunov-based adaptive controller is augmented that imposes state constraints to enforce strict adherence to the desired trajectory with predefined nominal bounds. By virtue of the constraint, the tracking error is bounded within a pre-defined bound. The bound is not time-varying because the reference trajectory is designed to begin from the current state of the robot with minimum jerk. The complete framework is demonstrated through hardware-in-the-loop experiments using a planar floating satellite platform and a prototype docking station. Experimental results corroborate the efficacy of autonomously achieving docking while satisfying stringent terminal constraints on position, velocity, and orientation, demonstrating the framework’s robustness and practical applicability to on-orbit servicing missions.

3D scene graph (3DSG) generation is a rapidly evolving field that plays a significant role in robotic autonomy. Traditionally, the focus has been on indoor environments, where robots understand and navigate spaces by abstracting objects and geometric information in a structured graph format. Expanding upon this idea, this paper introduces a 3DSG construction architecture, which enables scene-agnostic abstraction of the environment, with the goal of facilitating the adoption of 3DSG for autonomous agents in both indoor and outdoor environments. We propose a novel approach for area delimitation in 3DSGs that leverages label propagation to cluster entities (i.e. objects of interest) into areas that are both semantically and topologically distinguishable within a scene. Towards this end, we establish label propagation for 3DSGs, by formulating a dynamic set of propagation factors that accommodate to the relevance of semantic information and their natural decay through the topological structure of the 3DSG. Additionally, to achieve scene-agnostic area delimitation, we introduce a single-step optimization process for the calculation of clutter-aware propagation factors based on the approximation of an optimal set of factors that maximize inter-area eccentricity while minimizing intra-area eccentricity. Finally, the proposed framework is extensively validated through simulations and real-world deployments using a Boston Dynamics Spot legged robot and a Clearpath Husky mobile robot. The experimental results showcase the scalability of the proposed framework to indoor and outdoor environments for real-time 3DSG construction.

This article presents a novel coordination and task-planning framework to enable the simultaneous conflict-free collaboration of multiple uncrewed aerial vehicles (UAVs) for aerial 3D printing. The proposed framework formulates an optimization problem that considers a construction mission divided into sub-tasks and a team of autonomous UAVs, along with limited volume and battery. It generates an optimal mission plan comprising task assignments and scheduling, while accounting for task dependencies arising from the geometric and structural requirements of the 3D design, inter-UAV safety constraints, material usage and total flight time of each UAV. The potential conflicts occurring during the simultaneous operation of the UAVs are addressed at a segment-level by dynamically selecting the starting time and location of each task to guarantee collision-free parallel execution. An importance prioritization is proposed to accelerate the computation by guiding the solution towards more important tasks. Additionally, a utility maximization formulation is proposed to dynamically determine the optimal number of UAVs required for a given mission, balancing the trade-off between minimizing makespan and the deployment of excess agents. The proposed framework’s effectiveness is evaluated through a Gazebo-based simulation setup, where agents are coordinated by a mission control module allocating the printing tasks based on the generated optimal scheduling plan while remaining within the material and battery constraints of each UAV. A video of the whole mission is available at the following link: https://youtu.be/b4jwhkNPTyQ Note to Practitioners—This framework addresses the critical need for efficiency and safety in planning and scheduling multiple aerial robots for parallel aerial 3D printing. Existing approaches lack safety guarantees for UAVs during parallel construction. This work tackles these challenges by ensuring safety during parallel operations and effectively managing task dependencies. The framework incorporates material and flight time constraints for each UAV and determines the optimal number of UAVs required for a specific construction mission in a single computation step. Additionally, a task prioritization method is introduced, reducing the computational time of the optimization problem. This approach is particularly suited for applications such as rapid modular construction in remote or disaster-affected areas, where efficient UAV coordination is essential. The framework’s preliminary feasibility is demonstrated in a simulated environment, while the real-world experimentation is planned as future work.

This paper presents a novel framework based on edge computing, implemented using Kubernetes orchestration, to optimally offload the computational tasks required for centralized control of multiple robotic agents. Edge-based centralized control architectures are prone to failure due to communication delays. The proposed framework computes the maximum round-trip time delay for which the system remains stable and modifies the controller parameters to ensure the control computation within the critical time. For higher processing and communication delays, the complexity of the controller needs to be reduced by reducing the number of agents, the prediction horizon, and the efficient use of edge resources. The edge resources are dynamic, and the controller needs to be designed to guarantee the online computation within a desired time. A dynamic resource allocation method (based on an approximate function of the controller parameters, complexity, and computational resources) is proposed to design the controller parameters to ensure the bounded computation time. To validate the effectiveness of the proposed approach, we conduct experimental evaluations that analyze system behavior under various conditions, providing valuable insights into the performance, scalability, and robustness of multi-agent control systems deployed on edge infrastructure.

Reliable radar inertial odometry (RIO) demands precise correction of inertial measurement unit (IMU) acceleration drift during the synchronization of heterogeneous sensors. This bias drift, most evident in acceleration measurements, worsens in subterranean environments due to extreme cold or gravitational components acting along the robot’s feed-forward axis. Left uncorrected, such drifts substantially degrade sensor fusion performance, leading to dead-reckoning, particularly when employing cost-effective platforms such as the Pixhawk IMU in combination with multiple mmWave radars. In addition, radar point clouds differ fundamentally from LiDAR data, being inherently sparse, noisy, and prone to flickering effects, which further complicates the challenge of achieving stable and reliable odometry. To address these challenges, this article presents a novel two-stage multi-radar inertial odometry (MRIO) framework for resilient localization and mapping in GPS-denied subterranean environments. In the first stage, radar ego-velocity estimation, formulated through a least-squares approach, is incorporated into an Extended Kalman Filter (EKF) for IMU bias correction. The resulting drift-free accelerations are then fused in the second stage with measurements from multiple radars and the IMU to refine odometry performance. Beyond odometry, the framework also supports radar-based mapping by leveraging the generated robot’s translational and angular displacement by the proposed framework. Starting with a range-based outlier filter, least-squares ego-velocity reconstruction, the MRIO framework enables robust ego-velocity estimation, localisation, and mapping using only radar and IMU measurements. The proposed framework is extensively validated across multiple experimental scenarios, including real-time deployment in smoke-filled subterranean environments. Comparative evaluations involving multiple FMCW radar setups, LiDAR inertial odometry (LIO), and prior EKF-RIO methods further validate the framework’s robustness and its consistent capability to achieve precise localization and mapping in challenging perceptual environments.

The fifth generation (5G) cellular network technology has matured and is increasingly utilized in many industrial robotics applications. Various sectors seek to harness the advanced communication performance and deploy time-critical robotic applications in a connected manner, often hosting the execution of computational intensive time-critical components in the edge cloud. Robust deployment of cellular enabled robots that rely on the network performance demands the utilization of quality of service (QoS) solutions to respect the real-time requirements of critical applications. This paper proposes a method of harnessing the 5G QoS features in a dynamic fashion to retain the time-critical requirements of the edge-offloaded robotics applications. The paper emphasizes the dynamic selection of network resources considering the continuously changing communication requirements of such applications based on the underlying evolving mission and highlights the importance of deploying offloaded robotics applications to be communication aware — towards the era of co-design. Further, the dynamic nature of the background network traffic is examined and robot scalability with varying mission priorities is analyzed and discussed. A novel modeling approach coupling the time-critical performance of 5G-enabled robotics and the dynamic QoS selection is presented and utilized in the selection of the appropriate QoS profile. The lack of real-life evaluation of complete similar solutions is tackled by extensive experimental evaluation utilizing a real-life 5G stand alone (SA) network and a quadruped robot. The obtained results demonstrate the importance of such synergistic solutions.

Active Simultaneous Localization and Mapping (SLAM) has significantly enhanced the autonomy of robotic systems in real-time exploration, but there remains a constant pursuit for more adaptive and intelligent decision-making capabilities. This paper presents a hierarchical method for autonomous robotic navigation and mapping, leveraging frontier-based exploration and Deep Reinforcement Learning (DRL). The high-level frontier searching strategy provides waypoints for a low-level DRL-based path planner, enabling efficient and adaptive navigation in unknown environments. In the proposed methodology, once the robot reaches a desired waypoint, the high-level planner utilizes the current occupancy map created from mapping functions to identify the frontier with the highest entropy. This frontier is then set as the goal for the low-level policy, which is trained using deep reinforcement learning to generate a path to the goal while avoiding obstacles. Our approach has been implemented and evaluated using simulation environments in Gazebo and with real-time experiments. The results demonstrate that the hierarchical approach significantly enhances exploration decision-making capabilities.