CL-RaA*: Enabling fixed-wing UAV autonomous navigation in complex terrain
Tsinghua University Press
image:
The JITEP generates spatial trajectories that satisfy the differential constraints of fixed-wing UAVs only when needed. It does not require the pre-determination of multiple maneuver forms. Given the aircraft state at the current waypoint and the position of the successor waypoint, the JITEP utilizes the full-state six-DOF dynamics model and controller of the fixed-wing UAV to constrain the trajectory generation process, ensuring the uniqueness of the generated trajectory. Once the trajectory is successfully expanded, the aircraft state at the successor waypoint is added to the aircraft state tree. This ensures the continuity of the aircraft state at waypoints during the generation of the aircraft state tree.
view moreCredit: Chinese Journal of Aeronautics
Path planning, a pivotal constituent of Unmanned Aerial Vehicle (UAV) systems, has emerged as a prominent domain of inquiry, with research demonstrating its tangible applicability within the rotary-wing UAV domain. In contrast to their rotary-wing counterparts, fixed-wing UAVs offer a broader spectrum of utility owing to their heightened speed, extended range, and augmented payload capacity, facilitating applications spanning reconnaissance, surveillance, target tracking, aviation logistics, and aerial photography, among others. For the coverage problem in these tasks, current research has significantly improved mission completion efficiency. However, for fixed-wing UAVs, the inherent nonlinearity of their dynamic models and the complexity of their control mechanisms make ensuring the feasibility of the planned path a critical challenge.
Recently, an autonomous decision-making UAV team led by Yaoming Zhou from Beihang University, China, has reported a feasible path planning algorithm for fixed-wing UAVs termed Closed-loop Radial Ray A* (CL-RaA*). This work first proposes the Radial Ray A* (RaA*), a fast search algorithm based on an adaptive variable-step-size mechanism that guarantees rapid path searching even in environments with large concave obstacles. Additionally, to account for the dynamics and control characteristics of fixed-wing UAVs, a new expansion method, the Just-in-Time Expansion Primitive (JITEP), is introduced. By integrating JITEP with RaA* and under the constraint of safety checks, the CL-RaA* is further proposed. Finally, following comparative experiments with mainstream algorithms, a simulation platform for fixed-wing UAVs was constructed to validate the CL-RaA* algorithm in complex environments.
The team published their work in Chinese Journal of Aeronautics on May 3, 2025.
“In this work, a fast path search algorithm is first proposed, ensuring deterministic results while efficiently accommodating the UAV's motion characteristics across grid maps of different resolutions. Subsequently, a novel expansion method is introduced, which comprehensively accounts for the dynamics and control characteristics of a six-DOF UAV. By integrating these approaches and incorporating safety checks, we present the CL-RaA*, which is designed to generate safe and feasible paths more efficiently,”said Yaoming Zhou, the corresponding author of the paper, a professor in the School of Aeronautic Science and Engineering at Beihang University (China).
By properly setting the search step size, the RaA* algorithm can effectively balance the path length and run-ning time. Even in environments with large concave obstacles, increasing the step size can accelerate the search. In the eight characteristic environments, when the step size is set to 2, the average path length of RaA* increases by only 0.21% compared to the optimal path, while the search speed improves by a factor of 3.32. In the environment with random grid obstacles, the search speed of RaA* is approximately 5.76 times that of Goal-bias RRT and 28.75 times that of JPS. Compared to RRT-Dubins, RRT-JITEP, and RaA*-Dubins, the CL-RaA* achieves more effective feasible path planning for fixed-wing UAVs. Forty random tests in densely populated obstacle environments show that the CL-RaA* achieves an average trajectory length that is 42.19%, 22.01%, and 3.31% shorter than those of RRT-Dubins, RRT-JITEP, and RaA*-Dubins, respectively, and an average planning time that is reduced by 62.81%, 62.21%, and 4.76%, respectively. Compared to feasible trajectories generated using Dubins curves, those generated by the CL-RaA* algorithm are more conducive to UAV tracking. Test results in two random environments show that when tracking the feasible trajectories planned by CL-RaA* at a desired flight speed of 165 m/s, the maximum position error does not exceed 10 m, with a root-mean-square error of less than 3 m.
The next step is to integrate CL-RaA* with real-time onboard systems and extend it to cooperative path planning for UAV swarms.
Other contributors include Hui Gao, Yuhong Jia, Qingyang Qin and Liwen Xu from the School of Aeronautic Science and Engineering at Beihang University in Beijing, China.
Original Source
Hui GAO, Yuhong JIA, Qingyang QIN, Liwen XU, Yaoming ZHOU. Integrating just-in-time expansion primitives and an adaptive variable-step-size mechanism for feasible path planning of fixed-wing UAVs [J]. Chinese Journal of Aeronautics, 2025, https://doi.org/10.1016/j.cja.2025.103566.
About Chinese Journal of Aeronautics
Chinese Journal of Aeronautics (CJA) is an open access, peer-reviewed international journal covering all aspects of aerospace engineering, monthly published by Elsevier. The Journal reports the scientific and technological achievements and frontiers in aeronautic engineering and astronautic engineering, in both theory and practice. CJA is indexed in SCI (IF = 5.7, Q1), EI, IAA, AJ, CSA, Scopus.
Journal
Chinese Journal of Aeronautics
Article Title
Integrating just-in-time expansion primitives and an adaptive variable-step-size mechanism for feasible path planning of fixed-wing UAVs
Clustering optimization strategy for cooperative positioning system aided by UAV
image:
The main step of this method includes the clustering optimization and the locally-centralized CP. Firstly, each cluster broadcasts information to the UAV to determine which cluster is the HPC. After that, all of the clusters are reconstructed by utilizing the clustering optimization strategy. Based on these optimized cluster, three types of positioning-related information are utilized to achieve the locally-centralized CP.
view moreCredit: Chinese Journal of Aeronautics
Over the past few decades, leveraging the advancements in Vehicle-to-everything (V2X) technologies, CP techniques have emerged as a promising way to enhance the vehicle positioning performance by fusing positioning-related information from a group of participating vehicles travelling in dense urban environments. Especially, the CP performance could be further improved by introducing SRVs into CP networks, which is called SRV-aided CP. Unfortunately, due to the signal blockage and reflection in urban environments, CP systems may split into several sub-clusters that cannot be connected with each other in dense urban environments, in which the sub-clusters with few SRVs will suffer from degradation of CP performance. However, most of existing networking strategies mainly concentrate on integrating the positioning-related information within a cluster and overlooking some potential inter-cluster information. Thus, there is a clear demand for a more accessible, reliable approach to fully utilize inter-cluster potential information and improve the positioning performance of the whole CP systems.
Recently, a team of researchers led by Hongbo Zhao from Beihang University, China have introduced an Unmanned Aerial Vehicle (UAV) into the SRV-aided CP network and designed a locally-centralized CP method based on clustering optimization strategy for this CP network. This work not only provides a complete solution for inter-node communication, but also fully integrates the positioning resources of the entire network, aiming to improve the CP performance of SRV-aided networks.
The team published their work in the Chinese Journal of Aeronautics on May 28, 2025.
“In this work, we propose the clustering optimization strategy to fully integrate whole-net information and achieve data fusion for the CP system aided by UAV, using double differential Global Navigation Satellite System (GNSS) baseline and inter-node ranging measurements. Firstly, we formulate an inter-cluster communication scheme to achieve the information interaction. The UAV is utilized to communicate with high precision cluster and broadcast information to multiple low precision clusters. Then, we design the clustering optimization strategy, which cooperates high precision nodes with low precision sub-clusters, to achieve whole-net optimization. Finally, we utilize the locally-centralized factor graph optimization algorithm for each optimized cluster to complete data fusion.” said Hongbo Zhao, professor at School of Electronics and Information Engineering at Beihang University (China), a senior expert whose research interests include multi-node cooperative positioning and non-terrestrial communication network.
“Compared with the CP strategies that only consider the positioning-related information within a cluster, the strategy we proposed can make full use of inter-cluster potential available CP information and achieve higher CP performance. The simulation results show that the RMSE for the low precision cluster declines from 0.98 m to 0.40 m, and the RMSE for the high precision cluster also declines from 0.65 m to 0.22 m by using our method. These results indicate that the positioning accuracy and reliability of CoVs can be improved by utilizing available potential information from high-precision nodes.” said Hongbo Zhao.
Furthermore, Zhao put forward two major development directions may be pursued in future works. Firstly, Zhao intend to evaluate the impact of communication latency and positioning-related resource constraints to design a clustering optimization strategy for large-scale CP networks. Moreover, Zhao will consider to introduce more aerial cooperators into the CP system to further construct an aerial CP network, which can better assist ground nodes in inter-node communication, position awareness and path planning.
Original Source
Hongbo ZHAO, Zeqi YIN, Shan HU. Clustering optimization strategy for cooperative positioning system aided by UAV [J]. Chinese Journal of Aeronautics, 2025, https://doi.org/10.1016/j.cja.2025.103594.
About Chinese Journal of Aeronautics
Chinese Journal of Aeronautics (CJA) is an open access, peer-reviewed international journal covering all aspects of aerospace engineering, monthly published by Elsevier. The Journal reports the scientific and technological achievements and frontiers in aeronautic engineering and astronautic engineering, in both theory and practice. CJA is indexed in SCI (IF = 5.7, Q1), EI, IAA, AJ, CSA, Scopus.
Journal
Chinese Journal of Aeronautics
Article Title
Clustering optimization strategy for cooperative positioning system aided by UAV
The expanding horizons of INDI flight control
Tsinghua University Press
image:
Block diagram of a generic INDI controller with usual components
view moreCredit: Chinese Journal of Aeronautics
Traditional control systems like Nonlinear Dynamic Inversion (NDI) rely heavily on precise mathematical models. While this worked well for conventional aircraft, it falls short in today’s complex, fast-evolving platforms—especially those with novel configurations or uncertain dynamics. Inaccuracies in modeling can lead to instability, performance degradation, or complete system failure. Moreover, these systems often struggle to respond quickly to external disturbances or adapt in real time. With aerospace designs advancing and mission demands increasing, a shift toward more flexible, sensor-driven control strategies is critical. Due to these limitations, there is a pressing need to reimagine control architecture—and INDI has stepped into that gap.
In two open-access series survey papers (DOI: 10.1016/j.cja.2025.103553;10.1016/j.cja.2025.103591) recently published in the Chinese Journal of Aeronautics, a research team from the Technical University of Munich and Shanghai Jiao Tong University provides the most comprehensive review to date of Incremental Nonlinear Dynamic Inversion. Part I walks readers through the foundational logic and evolution of INDI—from model-based to sensor-based to hybrid systems—while Part II explores its modern extensions, including actuator-aware architectures and stability-enhancing components. Together, these studies serve as both a deep technical reference and a practical roadmap for the INDI-based flight control design.
Part I of the review traces INDI’s conceptual shift from model reliance to real-time adaptability. It introduces measurement-based INDI as a turning point: instead of calculating control inputs from models, the system reads real-world measurements to respond more quickly and robustly. This change allows aircraft to reject disturbances one step earlier than traditional methods—an edge in dynamic or unpredictable environments. Hybrid forms, blending model and sensor data, further enhance accuracy by adapting to flight conditions in real time.
The second part digs deeper into the components that make INDI work. It explains how reference models generate desired trajectories, error controllers guide corrections, and estimation modules synchronize commands with actuator states. New actuator-aware extensions—like Extended INDI and Actuator NDI—compensate for different effector dynamics, a vital feature in platforms like tilt-rotor drones or electric vertical take-off and landing (eVTOL) aircraft. The survey also reviews how INDI-based controllers allocate commands across multiple actuators without compromising system stability, even under saturation or time delay constraints.
Altogether, the two-part survey not only maps the technical terrain of INDI but also highlights its growing maturity, signaling that it is ready for large-scale deployment in aerospace and beyond.
“INDI marks a fundamental shift in how we think about control,” says Dr. Agnes Steinert, lead author of the survey. “Instead of designing around what we think the system should do, INDI lets us respond to what the system is actually doing in real time. That flexibility is essential for new aircraft architectures and uncertain environments. With this survey, we wanted to demystify the methodology and offer engineers a practical guide for integrating INDI into real-world systems.”
INDI offers a control approach that combines flexibility with inherent robustness, making it particularly well-suited for novel and increasingly complex flight systems such as tilt-wing drones, eVTOLs, and other emerging aircraft configurations. Its ability to adapt to changing dynamics and uncertainties in real time allows stable and reliable control even under challenging conditions where classical model-based methods face limitations. The sensor-based structure of INDI has also sparked interest in other fields, including marine and robotic systems, where similar challenges of modeling complexity and environmental disturbances arise. As the technology matures, INDI is steadily advancing from an academic concept toward practical integration in a wide range of safety-critical applications.
Original Source
Agnes STEINERT, Stefan RAAB, Simon HAFNER, Florian HOLZAPFEL, Haichao HONG. From fundamentals to applications of incremental nonlinear dynamic inversion: A survey on INDI – Part I [J]. Chinese Journal of Aeronautics, 2025, https://doi.org/10.1016/j.cja.2025.103553.
Agnes STEINERT, Stefan RAAB, Simon HAFNER, Florian HOLZAPFEL, Haichao HONG. Advancements in Incremental Nonlinear Dynamic Inversion and its Components: A Survey on INDI – Part II [J]. Chinese Journal of Aeronautics, 2025, https://doi.org/10.1016/j.cja.2025.103591.
About Chinese Journal of Aeronautics
Chinese Journal of Aeronautics (CJA) is an open access, peer-reviewed international journal covering all aspects of aerospace engineering, monthly published by Elsevier. The Journal reports the scientific and technological achievements and frontiers in aeronautic engineering and astronautic engineering, in both theory and practice. CJA is indexed in SCI (IF = 5.7, Q1), EI, IAA, AJ, CSA, Scopus.
Journal
Chinese Journal of Aeronautics
Article Title
From fundamentals to applications of incremental nonlinear dynamic inversion: A survey on INDI – Part I
No comments:
Post a Comment