Network Localization Based Planning for Autonomous Underwater Vehicles with Inter-Vehicle Ranging
This addresses localization reliability for AUV swarms in underwater operations, but it is incremental as it builds on existing sensor network localization techniques.
The paper tackles the problem of ambiguous localization in autonomous underwater vehicle (AUV) swarms due to low-rigidity network topologies by developing a rigidity-based planning and control framework that accounts for sensor noise and limited range. It shows the framework can generate feasible paths while guaranteeing minimum network rigidity in simulated 2D environments.
Localization between a swarm of AUVs can be entirely estimated through the use of range measurements between neighboring AUVs via a class of techniques commonly referred to as sensor network localization. However, the localization quality depends on network topology, with degenerate topologies, referred to as low-rigidity configurations, leading to ambiguous or highly uncertain localization results. This paper presents tools for rigidity-based analysis, planning, and control of a multi-AUV network which account for sensor noise and limited sensing range. We evaluate our long-term planning framework in several two-dimensional simulated environments and show we are able to generate paths in feasible time and guarantee a minimum network rigidity over the full course of the paths.