Guarded Motion and Reflexive Behaviors for the Survivor Buddy Robot

Abstract

This paper summarizes hardware and software changes made to the latest version of the Survivor Buddy, specifically regarding the addition of motor burnout prevention systems and the implementation of behavior tracking and mimicking functionality. The Survivor Buddy is an inexpensive and compact robot platform designed to interact with people as a social actor. Its main applications include search and rescue operations and telemedicine, with potential applications in other areas. Currently, robots in these areas are either very expensive, very large, or do not engage with people as social actors. The updates made to the Survivor Buddy outlined in this paper are designed to increase its ability to accurately model human behavior while allowing it to detect when its behavior-based movements are blocked. To create these changes, functionality from existing libraries was combined with data processing and classification algorithms. The implementation of the reflexive system created a system response time range of 0.5-1 seconds, allowing the robot to quickly detect obstacles. These changes demonstrate the efficacy of techniques designed to support inexpensive hardware, thereby providing proof of the practical feasibility of low-cost and functional social robot platforms.