Intelligent Sailing Model for Open Sea Navigation

TL;DR

The paper tackles the lack of realistic, interpretable interactive maritime traffic models for autonomous vessels. It introduces the Intelligent Sailing Model (ISM), a rule-adaptive framework that combines a high-level waypoint engine grounded in COLREGS with a low-level MPC-based controller, enabling open-sea navigation across multiple vessel types. Extensive numerical experiments show high goal-reaching rates and strong rule compliance in ism-only, mixed-traffic, AIS-derived, and multi-vessel scenarios, with scalable runtime performance. The work demonstrates that ISM can serve as a standard, efficient benchmark for evaluating and accelerating autonomous navigation algorithms in complex maritime environments, while offering clear interpretability and modularity for various vessel dynamics.

Abstract

Autonomous vessels potentially enhance safety and reliability of seaborne trade. To facilitate the development of autonomous vessels, high-fidelity simulations are required to model realistic interactions with other vessels. However, modeling realistic interactive maritime traffic is challenging due to the unstructured environment, coarsely specified traffic rules, and largely varying vessel types. Currently, there is no standard for simulating interactive maritime environments in order to rigorously benchmark autonomous vessel algorithms. In this paper, we introduce the first intelligent sailing model (ISM), which simulates rule-compliant vessels for navigation on the open sea. An ISM vessel reacts to other traffic participants according to maritime traffic rules while at the same time solving a motion planning task characterized by waypoints. In particular, the ISM monitors the applicable rules, generates rule-compliant waypoints accordingly, and utilizes a model predictive control for tracking the waypoints. We evaluate the ISM in two environments: interactive traffic with only ISM vessels and mixed traffic where some vessel trajectories are from recorded real-world maritime traffic data or handcrafted for criticality. Our results show that simulations with many ISM vessels of different vessel types are rule-compliant and scalable. We tested 4,049 critical traffic scenarios. For interactive traffic with ISM vessels, no collisions occurred while goal-reaching rates of about 97 percent were achieved. We believe that our ISM can serve as a standard for challenging and realistic maritime traffic simulation to accelerate autonomous vessel development.

Figures (4)

• Figure 1: ism integration with traffic environment. The depicted example consists of two ism vessels and one obstacle vessel that is not reacting to the other vessels. Based on the states $s_i$, the rule-adaptive waypoint engine generates the set of waypoints $\mathcal{W}$ that characterize the motion planning task and potentially collision avoidance maneuver. In this example, the ism vessel 1 is in a crossing situation at time step $t_1$ for which the waypoint $W_{c,1}$ is reached (see Fig. \ref{['fig:waypointadaption']}) and the goal waypoint is $G$. For the mpc, the path characterized by the waypoint set $\mathcal{W}$ is transformed to a desired trajectory $\mathbf{\xi}^d$. Then, the optimal control input $\mathbf{u}^*$ is computed to track the reference given the vessel dynamics $f$.
• Figure 2: Waypoint adaption for collision avoidance maneuvers of an ism vessel in head-on, crossing, and overtake situations as the give-way vessel. The obstacle vessel and its rule-compliant path are depicted in blue, and the ism vessel is in orange, with the desired avoidance path as a solid black line. Tunable parameters are gray.
• Figure 3: Trajectories and control inputs for three encounter situations: head-on, crossing, and overtaking (left to right). The marked time steps are as specified in Fig. \ref{['fig:waypointadaption']}. The upper row displays the trajectories in the position space with the optimal paths described by the waypoints in gray. The vessel is of type 1 and is depicted with its actual spatial dimension. The second and third rows show the angular velocity and normal acceleration over time, respectively.
• Figure 4: Multi ism vessel scenario with colored position trajectories and gray optimal paths for six vessels. The vessels are displayed in black with their proportional spatial dimensions at the initial time step and the time point 236s.