This paper presents a data-driven decentralized trajectory optimization approach for multi-robot motion planning in dynamic environments. When navigating in a shared space, each robot needs accurate motion predictions of neighboring robots to achieve predictive collision avoidance. These motion predictions can be obtained among robots by sharing their future planned trajectories with each other via communication. However, such communication may not be available nor reliable in practice. In this paper, we introduce a novel trajectory prediction model based on recurrent neural networks (RNN) that can learn multi-robot motion behaviors from demonstrated trajectories generated using a centralized sequential planner. The learned model can run efficiently online for each robot and provide interaction-aware trajectory predictions of its neighbors based on observations of their history states. We then incorporate the trajectory prediction model into a decentralized model predictive control (MPC) framework for multi-robot collision avoidance. Simulation results show that our decentralized approach can achieve a comparable level of performance to a centralized planner while being communication-free and scalable to a large number of robots. We also validate our approach with a team of quadrotors in real-world experiments.
I. INTRODUCTIONAutonomous navigation of a team of robots in dynamic environments is important when deploying them in various applications such as coverage and inspection [1], search and rescue [2], formation flying [3] and multi-view videography [4]. In these scenarios, the robots navigate in a shared space that may also have moving obstacles. To achieve predictive collision avoidance and ensure safety, each robot needs to know the future motion predictions of other robots in the environment. These motion predictions can be obtained among robots by sharing their future planned trajectories with each other via communication [5]. However, such communication may not be available nor reliable in practice. Alternatively, some approaches [6] employ a constant velocity model to predict other robots' trajectories. Even though communication among robots is not required, the planned robot motions may not be safe, particularly in crowded dynamic environments [5].In this paper, we propose an interaction-and obstacleaware trajectory prediction model and combine it with