Multi-robot teams learn to handle uncertain maritime tasks efficiently
Reinforcement Learning Based Heterogeneous Task Allocation Under Spatio-Temporal Environmental Uncertainty for Multi-Robot Teams
Robotics
Summary
Maritime robots called Autonomous Surface Vehicles (ASVs) can do many different tasks, but unpredictable conditions like waves and wind make deciding who does what tricky. The authors teach a group of these robots to decide on tasks by themselves using a learning method that helps them handle surprises in the environment. Their approach lets the robots work well together to finish many different kinds of tasks faster and better than other methods. They tested this in computer simulations with up to 100 robots and 400 tasks.
What this means in practice
- •For maritime robotics teams: Automatically assign tasks to diverse ASVs under changing environmental conditions to complete missions faster and more reliably.
- •For industrial automation engineers: Implement decentralized control policies for teams of different robots performing multiple task types in uncertain environments.
Authors
Xi Lin, Yaxuan Li, Paul Szenher, Brendan Englot
Abstract
Autonomous Surface Vehicles (ASVs) have been increasingly deployed to perform different kinds of maritime tasks in recent years, which motivates the need for a reliable task allocation policy that can adapt to heterogeneous tasks and maritime disturbances. In this work, we formulate a multi-robot task allocation problem in which a team of ASVs with heterogeneous capabilities needs to complete tasks of different types and appearance times under unknown stochastic spatio-temporal uncertainties from maritime disturbances. We propose a decentralized Reinforcement Learning (RL) based policy network that is trained to be robust towards uncertainty-induced observation deviations, and is able to perform individual decision making that leads to efficient execution behaviors toward heterogeneous task requirements. We evaluate the proposed policy and four baseline methods in simulated experiments with varied proportions of different tasks types, numbers of tasks, robot species and robots per species. Experiment results show that our proposed policy achieves the lowest makespan and can be deployed to address scenarios with 100 robots and 400 tasks.