Team 44
Team Members |
Faculty Advisor |
Matthew Ghanie |
Wei Wei Sponsor Movia Robotics |
sponsored by
Sponsor Image Not Available
Robot Control through Participant Position and Posture Detection for Human Robotic Interaction
Create a system that enables a robot to mimic the actions of a child. System for capturing segmenting and characterizing body posture and movement. The system uses an optical camera mounted in the head of the robot to capture participants in front of the robot. By utilizing skeleton localization libraries to analyze the imagery we can extract the posture of a person. The intention is to build a system that can first extract body position, posture and gestural movements then provide real-time information of the participants body part’s position relative to the robot for immediate use to alter the posture of the robot, and finally store them for later use in other processes. A list of deliverables and requirements follows below. Since the camera can be moved the process of detecting the position of the participant will incorporate the relative orientation and position relative to the robot center of axis. The goal will be to detect three main parameters of the participant. 1. The overall position of the participant relative to the robot. 2. The position and orientation of the participants head and face. 3. The overall posture and gestures of the participants’ torso, head and arms.