Enhancing Navigational Safety in Crowded Environments using Semantic-Deep-Reinforcement-Learning-based Navigation
This work addresses safety challenges for mobile robotics in applications like delivery and healthcare, but it is incremental as it builds on existing deep reinforcement learning methods by adding semantic information.
The paper tackles the problem of safe navigation in crowded, dynamic environments with diverse obstacles by proposing a semantic deep reinforcement learning approach that incorporates object-specific safety rules and danger zones. The result is an increase in safety compared to a benchmark method, with the agent learning to adjust safety distances based on semantic information.
Intelligent navigation among social crowds is an essential aspect of mobile robotics for applications such as delivery, health care, or assistance. Deep Reinforcement Learning emerged as an alternative planning method to conservative approaches and promises more efficient and flexible navigation. However, in highly dynamic environments employing different kinds of obstacle classes, safe navigation still presents a grand challenge. In this paper, we propose a semantic Deep-reinforcement-learning-based navigation approach that teaches object-specific safety rules by considering high-level obstacle information. In particular, the agent learns object-specific behavior by contemplating the specific danger zones to enhance safety around vulnerable object classes. We tested the approach against a benchmark obstacle avoidance approach and found an increase in safety. Furthermore, we demonstrate that the agent could learn to navigate more safely by keeping an individual safety distance dependent on the semantic information.