Classical path planning algorithms such as A* applied to multi-UAV systems are unable to cope with multi-UAV complex scenarios in urban low altitude. This is due to the limitations in computational speed and convergence at a local optimal which are unable to guarantee the optimal flight energy and mission success rate. For such purposes, the Upgraded Gooseneck Barnacle Optimization (UGBO) algorithm is proposed based on the concept of swarm intelligence meta-heuristic intelligence with three mechanisms of improvements. They are opposing-based learning for better diversity of initial population, self-adaptive population adjust to balancing explore and exploit, and a forbidden strategy to get out of the local optimal. An improved energy consumption model considering aerobic drag, and wind impacts are employed, with clear multi-UAV coordination in the process through horizontal layer and proximity deconfliction in three-dimensional space, so the trajectory will be physically realizable and collision-free. Simulation in a 100 m × 100 m × 50 m simulation environment including fifteen to twenty static obstacles, and three to five dynamic obstacles show the UGBO achieve the path length nine hundred fifty meters, flight energy four hundred eighty units, calculation time twelve seconds, and average optimal fitness 0.85. These are, compared to other three classical algorithms A*, ACO, and PSO, improved by 19.5-33.1%, 19.3-33.3%, and 25.0-112.5%, respectively. Also, comprehensive experimental comparison with the state-of-the-art research such as PR-DQN, APF-RRT, T-DRL, and AHMP demonstrates that UGBO perform well with 3.1-17.4% improvement from those four approaches. Through ablating test, the study show that the improvement of opposition-based learning plays the most significant role in all three mechanisms. The results show an efficient multi-UAV path planning for urban aerial vehicles that outperform current solutions and can be readily integrated into learning-based approaches and deployed on physical hardware.
Single-Robot Motion Planning (SRMP) in highly non-convex constrained environments, where robots must satisfy collision-free guarantees, dynamic feasibility, and task-related constraints, is challenging under complex constraints and computational limits. Recent Model-Based Diffusion (MBD) approaches recast the SRMP as trajectory optimization that samples from a posterior over trajectories, using known dynamics, and analytically estimates the score function from rollout samples to guide diffusion denoising toward a low-cost, clean trajectory without demonstration learning. While existing works further adapt MBD to constrained environments and showcase promising performance, they are still limited by (1) enforcing safety either via soft feasibility diffusion priors or hard projection operators, but lack a unified framework to integrate both, and (2) fixing safety enforcement to neglect the changing of diffusion scheduling. Therefore, we introduce Model-Based Diffusion via Constraint Optimization and Adaptive Scheduling (MD-COAS) for SRMP that unifies the inexact Augmented Lagrangian Method (iALM) soft diffusion prior with a Convex Feasible Set (CFS)-based hard projection operator, and adaptively schedules and co-optimizes safety enforcement, along with diffusion scheduling. Experiments demonstrate that our method achieves higher safety \&success rates, faster convergence, and lower final costs than baseline planners on randomly generated highly non-convex 2D benchmarks and a 7-DoF robot arm avoidance task.
Zhilin He, Bowei Li, Jianlin Dou et al.· 1 citation
Multi-Robot Motion Planning in continuous environments, where robots must generate dynamically feasible, collision-free trajectories, is challenging due to the combinatorial growth of the joint trajectory space and the difficulty of enforcing dynamic feasibility and hard safety constraints. Recent approaches recast trajectory planning as probabilistic inference, sampling from a posterior over trajectories using diffusion models whose score functions are learned from demonstration data. While showing promising performance, these approaches are limited: they often rely on sizable demonstration datasets and struggle to rigorously enforce dynamics and hard safety constraints during sampling. To this end, we introduce Model-Based Diffusion Optimal Control (MDOC), a model-based diffusion planner that efficiently produces dynamically feasible trajectories without relying on data. Crucially, we show that MDOC's safety mechanism -- combining known dynamics models with Control Barrier Function-constrained projections -- naturally scales to multi-robot planning settings through Conflict-Based Search. Across simulation experiments, this integrated method consistently outperforms representative baseline planners in sample efficiency, geometric smoothness, and success rate, while reducing computation time and producing collision-free trajectories.