Learning to Navigate in Indoor Environments: from Memorizing to Reasoning
This work addresses the problem of map-free navigation for mobile robots, offering a solution that generalizes to unseen targets, though it is incremental as it builds on existing deep reinforcement learning methods.
The paper tackles autonomous navigation for mobile robots in indoor environments without requiring a pre-existing map, using deep reinforcement learning to enable robots to reach new, unseen goals with only RGB images and odometry, achieving successful navigation in both simulated and real-world tests.
Autonomous navigation is an essential capability of smart mobility for mobile robots. Traditional methods must have the environment map to plan a collision-free path in workspace. Deep reinforcement learning (DRL) is a promising technique to realize the autonomous navigation task without a map, with which deep neural network can fit the mapping from observation to reasonable action through explorations. It should not only memorize the trained target, but more importantly, the planner can reason out the unseen goal. We proposed a new motion planner based on deep reinforcement learning that can arrive at new targets that have not been trained before in the indoor environment with RGB image and odometry only. The model has a structure of stacked Long Short-Term memory (LSTM). Finally, experiments were implemented in both simulated and real environments. The source code is available: https://github.com/marooncn/navbot.