Abstract:In this paper, a swarm trajectory-planning method is proposed for multiple autonomous surface vehicles (ASVs) in an unknown and obstacle-rich environment. Specifically, based on the point cloud information of the surrounding environment obtained from local sensors, a kinodynamic path-searching method is used to generate a series of waypoints in the discretized control space at first. Next, after fitting B-spline curves to the obtained waypoints, a nonlinear optimization problem is formulated to optimize the B-spline curves based on gradient-based local planning. Finally, a numerical optimization method is used to solve the optimization problems in real time to obtain collision-free, smooth and dynamically feasible trajectories relying on a shared network. The simulation results demonstrate the effectiveness and efficiency of the proposed swarm trajectory-planning method for a network of ASVs.
What problem does this paper attempt to address?
The paper primarily addresses the issue of distributed swarm trajectory planning for multiple Autonomous Surface Vehicles (ASVs) in complex and obstacle-dense marine environments.
Specifically, the paper presents the following contributions:
1. **Multi-objective Optimization Problem Construction**: Unlike existing trajectory planning methods for a single ASV or multiple Unmanned Surface Vehicles (USVs), this paper constructs a multi-objective optimization problem based on local sensor information, which includes optimal energy consumption, dynamic feasibility, curve smoothness, obstacle avoidance distance, and endpoint distance constraints. This approach enables ASVs to complete autonomous navigation in unknown complex environments relying solely on onboard local sensor information, without the need for external positioning, computational support, or pre-constructed maps.
2. **Phased Planning Strategy**: The planning method decomposes the optimal trajectory generation problem into two stages: initial trajectory generation and backend trajectory optimization. First, a heuristic-based kinodynamic path search algorithm is used to efficiently find a safe, feasible, and time-efficient initial path. Subsequently, the initial path is optimized using a B-spline-based optimization combined with a gradient-based planning method, ensuring that the generated trajectory meets dynamic feasibility and enhances safety.
3. **Real-time Environment Perception and Global Trajectory Formation**: By utilizing point cloud information obtained from local sensors, ASVs can perceive the surrounding environment in real-time and generate real-time locally optimal trajectories accordingly. Additionally, this work uses multiple locally optimal trajectories to form a global trajectory, better meeting real-time obstacle avoidance requirements and generating shorter navigation distances for short-range autonomous navigation.
In summary, this research aims to address the key issue of achieving distributed swarm trajectory planning for multiple ASVs in complex and obstacle-dense environments. By proposing a novel planning method, it effectively improves the autonomous navigation performance and safety of ASV swarms.