Secure High-Bandwidth Communications for a Fleet of Low-Cost Ground Robotic Vehicles GOALS. The proposed research shall focus on meeting critical objectives toward achieving the long-term goal of developing a fleet of Flexible Autonomous Machines operating in an uncertain Environment (FAME). While the development of the fleets is well on its way (see recent detailed 377 page MS thesis supervised by Dr. Rodriguez [1]), a critical component that is missing is inter-vehicle communication. The proposed research shall therefore focus on providing secure high-bandwidth communications between fleet vehicles. MOTIVATION. The above goal is worthwhile because it will permit fleet vehicles (currently 7 vehicles) to perform much more complicated coordinated maneuvers than they can at the moment [1]. For example, suppose that the current vehicles are traversing a figure 8 planar track; i.e. a track where vehicles criss cross! Suppose that the vehicles have been commanded to traverse the track in minimum time and collision avoidance has been activated [2], [3]. Even with collision avoidance activated, a minimum time constraint can result in severe collisions. To avoid this, vehicles should communicate with one another ( signal ahead ) to let each other know when they are simultaneously approaching an intersection. With such look-ahead signaling, simple intersection protocols can be used to ensure a rapid safe flow of traffic. Now consider a platoon of vehicles in which spacing control has been activated [2]. In order to get tight spacing control for the platoon in the presence of an accelerating leader, it is useful for the leader to broadcast its acceleration profile to following vehicles [4]-[5]. More generally, inter-vehicle communication can be particularly useful for applications in which fleet vehicles are trying to optimize the collective performance of the fleet; e.g. search and rescue, intelligence gathering, mapping of an unknown environment [6], etc.. OBJECTIVES. The paramount objectives of the proposed work are (1) high-bandwidth communications between fleet vehicles and (2) security with respect to jamming, noise, and aggressive cyber attacks. CRITICAL QUESTIONS. Toward the above objectives, the following critical questions will be addressed during the proposed work period (Spring, 2016): (1) High-Bandwidth Communications. How can we ensure high-bandwidth communications between fleet vehicles? Here, we would like to transmit inertial measurement unit (IMU) data (e.g. speed, angular speed), ultrasonic sensor sate (e.g. distance from an obstacle), and video. The latter, particularly, will require a high-bandwidth link. A video link has been demonstrated within [1] using WiFi. (2) Multi-Rate Sampling to Maximize Bandwidth. In an effort to maximize bandwidth, how to we systematically implement multiple sampling rates; i.e. sampling at a lower rate when the spectral content of the signal permits it. Toward this end, the sample-data ideas within [7] are of great value. As long as we have a common period within our sampled-data system, the ideas within [7] can be extended directly to the mulita-rate case [8]. This is very powerful because it will permit
us to answer questions like how slow can we sample a signal without sacrificing stability or performance? (3) Vulnerability Assessment. Given the above, what are critical vulnerability points? That is points where an external cyber attacker can remotely infiltrate to take control and do damage. Here, we shall become very familiar with different vulnerability assessment and management tools [9]. (4) Vulnerability Protection. Once fleet/network vulnerabilities have been assessed, the next natural question to ask is how can we introduce a protective layer of software and/or hardware in order to remove (or significantly reduce) system vulnerabilities? Here, we shall be looking for a costeffective solution. BROADER IMPACT. To properly assess the broader impact of the proposed work, it is essential to note the following: The proposed work will permit optimization of overall fleet performance. As such, it will significantly elevate the existing fleet to a state whereby it represents a very powerful testbed for substantive and cutting-edge FAME research. Moreover, the fleet can be used as a very effective outreach tool. Given this, the potential impact of the proposed project can be very significant particularly when one factors the Arizona-wide impact that Dr. Rodriguez Engineering Academy efforts will have on students across Arizona; particularly low-income students, woman and underrepresented minorities. In short, students already find the capabilities of our fleet vehicle to be very compelling; e.g. cruise control, position control, ability to follow a curve in space, multiple vehicles following a leader with spacing control, minimum time around a racetrack, vision-based positioning, obstacle avoidance, cameralidar based environment mapping, camera-based localization [1]-[3], [6], [10]. The proposed work shall take our current capabilities to new level. The planned security work can also help shed light on a global problem that is accelerating and requires a new generation of trained engineers to meet the anticipated challenges. Final Demonstrations and Dissemination. The proposed final demonstration will demonstrate multiple vehicles traversing a criss crossing racetrack as discussed above. Advanced signaling can resolve collisions at intersection. This, of course, has very broad commercial applications. We shall also demonstrate multi-vehicle xy-planar coordination in order to optimize some xy-plane performance measure; e.g. minimum time to assume a minimum vulnerability fleet configuration. All results shall be documented in a final comprehensive report that will serve as the basis for writing a paper that will be submitted to the IEEE Transactions on Control Applications.
REFERENCES [1] Z. Lin, Modeling, Design and Control of Multiple Low-Cost Robotic Ground Vehicles, ASU MS Thesis, Electrical Engineering, (Supervisor: Dr. A.A. Rodriguez), 377 pages, August, 2015. [2] A.A. Rodriguez, Z. Lin, J. Aldaco, N. Ravishankar, K. Puttannaiah, Modeling, Design and Control of Multiple Low-Cost Differential-Drive Robotic Ground Vehicles, submitted for publication in the Proceedings of the 2016 American Control Conference (ACC), 8 pages. [3] J. Aldaco, Use of Vision and Optimal Control for Fleets of Ground Vehicles, ASU MS Thesis, December 2015. [4] S.E. Sheikholeslam, C. Desoer, Longitudinal Control of a Platoon of Vehicles, American Control Conference, IEEE, pp. 291-296, 1990. [5] S.E. Sheikholeslam, Control of a Class of Interconnected Nonlinear Dynamical Systems: The Platoon Problem, University of California, Berkeley, PhD Thesis, 1991. [6] X. Lu, Use of Lidar for Mapping Unknown Environments with Mobile Rear-Wheel Drive Ground Vehicles, ASU MS Thesis, Spring 2016. [7] T. Chen and B. Francis, Optimal Sampled-Data Control Systems, Springer Verlag, 1995. [8] K. Puttannaiah and A.A. Rodriguez, Multi-Rate Control of Multivariable Linear Time Invariant (LTI) Dynamical Systems, in preparation, 2015. [9] P. Phatak, Top 10 Security Assessment Tools, Open Source, Feb 22, 2012. http://opensourceforu.efytimes.com/2012/02/top-10-security-assessment-tools/ [10] R. Nikhilesh, Utility of a Camera-Lidar-Based Robot Localization System: Command, Control and Tracking Applications, Arizona State University, MS Thesis, December 2015.
TIMELINE FOR Secure High-Bandwidth Communications for a Fleet of Low-Cost Ground Robotic Vehicles Semester: Spring ZZZ Comprehensive Literature Survey October 2015-Jan 2016 Build 2 Enhanced Differential-Drive Robots January 2016 High-Bandwidth Communications January- February 2016 Vulnerability Assessment February-March 2016 Vulnerability Removal/Minimization March-April 2016 Collective Fleet Performance Data Collection April 2016 Final Data Collection, Demonstration, Final Report April-May 2016
BUDGET FOR Secure High-Bandwidth Communications for a Fleet of Low-Cost Ground Robotic Vehicles Semester: Spring 2016 2 Enhanced Thunder Tumbler Robotic Vehicles $315.32 Books, Supplies, Other Hardware $ 84.68 TOTAL: $400.00 Cost Breakdown for Enhanced Thunder Tumbler Robot Product/Component Quantity Price ($) $ Thunder Tumbler Vehicle 1 $10 $ Raspberry Pi 2 Model B 1 $40 $ Arduino Uno 1 $ $12.19 Adafruit Motor Shield 1 $ $22.50 Raspberry Pi 5MP Camera 1 $25 $ Camera Holder 1 $ $5 HCSR04 Ultrasonic Sensor 1 $ $1.87 Power Supply for Raspberry Pi 1 $10 $ Power Supply for Arduino 4 $ $6.75 Magnetic Wheel Encoders 2 $ $4.40 Adafruit 9dof IMU 1 $ $19.95 Total Price $ $157.66 Two Enhanced Thunder Tumblers will be built so that we can progress toward a fleet size of approximately 16 vehicles.