Technical Report: Reactive Semantic Planning in Unexplored Semantic Environments Using Deep Perceptual Feedback
This work addresses the challenge of autonomous navigation and interaction in dynamic, semantically rich environments for robotics applications, representing an incremental improvement by combining existing techniques like object detection and semantic SLAM.
The paper tackles the problem of reactive planning in unexplored environments by integrating semantic representations with deep perceptual learning, enabling real-time responses to human motions and gestures while maintaining collision avoidance. It demonstrates empirical utility through numerical comparisons with a state-of-the-art algorithm and physical implementations on wheeled and legged platforms.
This paper presents a reactive planning system that enriches the topological representation of an environment with a tightly integrated semantic representation, achieved by incorporating and exploiting advances in deep perceptual learning and probabilistic semantic reasoning. Our architecture combines object detection with semantic SLAM, affording robust, reactive logical as well as geometric planning in unexplored environments. Moreover, by incorporating a human mesh estimation algorithm, our system is capable of reacting and responding in real time to semantically labeled human motions and gestures. New formal results allow tracking of suitably non-adversarial moving targets, while maintaining the same collision avoidance guarantees. We suggest the empirical utility of the proposed control architecture with a numerical study including comparisons with a state-of-the-art dynamic replanning algorithm, and physical implementation on both a wheeled and legged platform in different settings with both geometric and semantic goals.