Detecting Points of Interest for Robotic First Aid
|Title||Detecting Points of Interest for Robotic First Aid|
|Summary||Detecting Points of Interest for Robotic First Aid|
|Keywords||Robot, First Aid, Visual Recognition|
|References|| -pose recognition
Jamie Shotton, Ross Girshick, Andrew Fitzgibbon, Toby Sharp, Mat Cook, Mark Finocchio, Richard Moore, Pushmeet Kohli, Antonio Criminisi, Alex Kipman, Andrew Blake, "Efficient Human Pose Estimation from Single Depth Images", IEEE Transactions on Pattern Analysis & Machine Intelligence, vol.35, no. 12, pp. 2821-2840, Dec. 2013, doi:10.1109/TPAMI.2012.241
Travers, A. H., Rea, T. D., Bobrow, B. J., et al. (2010). Part 4: CPR overview 2010 American Heart Association guidelines for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation, 122(18 suppl 3), S676-S684.
|Prerequisites||some capability to work with software, and interest in visual recognition and robots|
|Supervisor||Anita Sant'Anna, Martin Cooney|
Robots will save lives by conducting first aid in homes and public places when medical experts are unavailable. This project focuses on the case of an unconscious fall victim in a home. First aid, as described by the acronym CABD, can involve chest compressions, airway adjustments, artificial respiration, and treatment for bleeding.
Goal: the capability for a robot to locate points of interest for first aid: the sternum for chest compressions, the chin for the airway, the mouth for breathing, and points of possible bleeding.
Relation to some previous work:
A previous project by Tianyi Zhang and Yuwei Zhao involved a robot assessing some simple factors related to a fallen person's health state assuming depth cameras where placed directly above a person. The current project does not require this assumption. Another previous project involved a robot sent by an intelligent environment to ask a conscious fall victim if they were okay, but could not recognize a person's pose to conduct first aid. The current project will allow both of these works to be brought together.
- a robot will move to acquire visual data of a fallen person using a camera located on its arm
- visual data will be processed to find points of interest for first aid
- the robot will somehow indicate the points it has found
- Becoming (more) familiar with OpenCV, ROS, Arduino
- Basic literature review
- Getting robot and robot arm to move
- Acquiring some data and labelling
- Building a recognition system (the focus of the project)
Evaluation metric: Distance of output locations of points of interest and ground truth (Time until finding points.)
Focus: software (visual recognition)
Expected results: a thesis/report, code, video, (if possible a six page shortened version of the thesis, to be submitted to a conference)