Towards Target-Driven Visual Navigation in Indoor Scenes via Generative Imitation Learning

Abstract

We present a target-driven navigation system to improve mapless visual navigation in indoor scenes. Our method takes a multi-view observation of a robot and a target as inputs at each time step to provide a sequence of actions that move the robot to the target without relying on odometry or GPS at runtime. The system is learned by optimizing a combinational objective encompassing three key designs. First, we propose that an agent conceives the next observation before making an action decision. This is achieved by learning a variational generative module from expert demonstrations. We then propose predicting static collision in advance, as an auxiliary task to improve safety during navigation. Moreover, to alleviate the training data imbalance problem of termination action prediction, we also introduce a target checking module to differentiate from augmenting navigation policy with a termination action. The three proposed designs all contribute to the improved training data efficiency, static collision avoidance, and navigation generalization performance, resulting in a novel target-driven mapless navigation system. Through experiments on a TurtleBot, we provide evidence that our model can be integrated into a robotic system and navigate in the real world. Videos and models can be found in the supplementary material.

accepted by IEEE Robotics and Automation Letters

References in corpus (7)

• Spectral Normalization for Generative Adversarial Networks
• On Evaluation of Embodied Navigation Agents
• Reinforcement Learning with Unsupervised Auxiliary Tasks
• Reinforcement Learning from Imperfect Demonstrations
• Learning and Querying Fast Generative Models for Reinforcement Learning
• CrowdMove: Autonomous Mapless Navigation in Crowded Scenarios
• Improving Target-driven Visual Navigation with Attention on 3D Spatial Relationships