Intel Linux Security Robot

Conclusion

The outcome of this project demonstrates that it is possible to produce a security robot, or any robot, using nonspecific hardware and open source software. This is contrary to the recent trend toward the mindset that design systems necessitate design specific hardware and software not produced by an project's internal developers must cost money.  It is our hope that we have demonstrated this is no longer the case.

As an embedded computer, the core of the security robot consumes very little power.  Comparable video monitoring systems consume significantly more power than the robot. Moreover, maintenance costs and power requirements of the robot are negligible as compared to the salary of one or more human security personal capable of performing the same duties.  For any corporation or institution, it is the most financially sound choice in reliable security.

Due to the number of sensors that were utilized, the AI was not as responsive as desired. Data handling would take too long and lag the AI processing.  Another setback was that the robot's overall power consumption was much higher than we had originally anticipated. The 12 Volt battery was only capable of operating the robot for roughly 20 minutes at a time, requiring a constant stream of inconvenient recharging.

Future Direction

Future direction would be to implement additional sensor-enabled functionality, to improve the adaptability of the AI algorithm, and to work on improving the robot's total power consumption. Null Spark utilized sonar, thermopile, and motion capture webcamera hardware. While the sonar sensors were able to capture large distances, an IR range finder would have been more precise. Bump and photovoltaic sensors could also have been used to improve the robot's ability to perform security functions.

Recommendations include purchasing additional parts to accommodate for the possibility of hardware failure and creating detailed test cases for all desired functionalities.

 

Special Thanks

To Dr. Yinong Chen, our project advisor and Intel Corp. for all the support and encouragement they have given us these past two semesters. Without their guidance, we would not have been able to accomplish as much as we had.