MAKER: Development of Smart Mobile Robot System to Help Middle School Students Learn about Robot Perception

Transcription

1 Paper ID #14537 MAKER: Development of Smart Mobile Robot System to Help Middle School Students Learn about Robot Perception Dr. Sheng-Jen Tony Hsieh, Texas A&M University Dr. Sheng-Jen ( Tony ) Hsieh is a Professor in the Dwight Look College of Engineering at Texas A&M University. He holds a joint appointment with the Department of Engineering Technology and the Department of Mechanical Engineering. His research interests include engineering education, cognitive task analysis, automation, robotics and control, intelligent manufacturing system design, and micro/nano manufacturing. He is Director of the Rockwell Automation laboratory at Texas A&M University, a stateof-the-art facility for education and research in the areas of automation, control, and automated system integration. He also serves as Director of an NSF Research Experiences for Teachers (RET) program in the area of Mechatronics, Robotics, and Industrial Automation. Mrs. Vania Willms, c American Society for Engineering Education, 2016

2 Teaching Robot Perception in Middle School Abstract Robots are key to manufacturing, healthcare, entertainment, and aerospace exploration feature. The industry is in great need of qualified professionals that can meet the demand of the everchanging technologies and latest innovation. Robot perception is greatly researched and in demand on most fields in this industry. Blending these subjects in the classroom can be expended to motivate students to pursue careers in Science, Technology, Engineering, and Math (STEM). Image and video capture using a cell phone camera and VEX sensors can be explored into more depth in the middle school classroom. This study examined the use of VEX sensors and an iphone 6 camera as an introduction to robot perception to middle school students. VEX line followers, ultrasonic rangefinders, and an iphone camera were used to perform object recognition and conduct robot navigation within a classroom robotics competition field setting. Overview Computer science drives innovation and is one of the fastest growing fields in our economy, opening doors to high paying jobs. In addition, programming is an important skill regardless of career choice. When students learn to program, they also learn important problem-solving, creative thinking, and computational thinking skills. Robots provide a hands-on activity that directly involves students in the learning process and allows them to take ownership of their STEM learning. Robots bring code to life and allow students to see how what they re learning has a direct impact in the real world, and how individual math and engineering elements come together to form a solution to a real problem. RobotC uses the industry standard C-programming language. Students learn and practice the same type of programming that is used in advanced education and professional applications. By engaging students in programming robots, their interest in programming can be fostered, allowing them to transition to more advanced programming such as object recognition. This study examined the use of VEX sensors and an iphone 6 camera as an introduction to robot perception to middle school students. VEX line followers, ultrasonic rangefinders, and an iphone camera were used to perform object recognition and conduct robot navigation within a classroom robotics competition field setting. Learning materials were created to help middle school students learn to build a robot and train it to perceive objects in 6 weeks. Lesson Design Engineering Connection Part of the engineering design process is to design effective solutions. In order to develop curious critical thinkers, it's important to teach them that learning possibilities are endless, and failure is a natural part of the process. Engineers and Scientists design, test, and redesign solutions until

3 their objective is achieved, or a new discovery is made. In this lesson, students will learn how to calculate thresholds and program line tracking and sonar sensors that will be used to design solutions for an exciting challenge. Introduction/Motivation Show student Machine vision capabilities enable autonomous navigation, human robots the Autonomous Robot Soccer video: Activities Students were given a task to research computer and machine vision. Second, they designed and built their robot utilizing ultrasonic range finder sensors to enable autonomous mode. Third, they used an iphone application to send images to the computer. Finally, they used simple algorithms to process the images learning how computer vision works. The programming for the robot autonomous mode was done on RobotC. The programming for image Processing was done on XCode using OpenCV libraries. Materials Item Quantity Cost VEX Clawbot Kit 1 $ Ultrasonic Range Finder 4 $29.99 Line Tracker 1 $39.99 iphone 6 1 $ TOTAL $ Student Groups The class was divided into four groups. One team did research on streaming video; as second team worked on designing and building the robot; a third worked on the literature review, presentation, and media; and the fourth did research on object recognition. Hardware VEX Clawbot Kit The VEX Clawbot Kit includes the basic robot for the VEX EDR robots. The kits included all parts necessary for assembling the robot as it will work with the object perception robot: 300+ parts including 4 motors, structural metal, fasteners, wheels, and gears. Ultrasonic Range Finder An ultrasonic range finder sensor enables a robot to detect obstacles in its path by utilizing the propagation of high-frequency sound waves. The sensor emits a 40kHz sound wave, which

4 bounces off a reflective surface and returns to the sensor. Then, using the amount of time it takes for the wave to return to the sensor, the distance to the object can be computed. The ultrasonic range finder emits a high-frequency sound wave that alerts the robot to things in its path. A Programming Kit is needed to change the program in the VEX Controller. These are specific behaviors achieved by the ultrasonic range finder: measure distances from 1.5in to 115in; detect obstacles using high frequency sound waves; create more autonomous functions. The sensor can be used to determine distances to objects. It can be used as a tool to determine if any objects are in the robot s path at all. To increase the sensing range, the sensor can be mounted to a servo to allow it to rotate. The sensitivity of the sensor depends on the objects surfaces that are detected by the emitted sound waves. For example, a reflective surface may produce a different reading than a nonreflective surface. The resolution of the sensor also depends on the sound waves. However, sound waves can reflect or be absorbed and possibly not return with enough power. Sensitivity: Detect a 3cm diameter pole at greater than 2m. Usable Range: 3.0 centimeters meter / 1.5 inches inches Frequency: 40 KHz Resolution: 1 inch

5 Figure 1 - Beam pattern/angle Line Tracker The line tracker allows the robot to follow a black line over a white surface. This kit provides more design flexibility-. It can also be used to program the robot for light sensing. A Programming Kit is needed to change the program in the VEX Controller. These are specific behaviors a liner tracker enables a robot to perform: navigate down a marked path; use for light sensing; more autonomous functions. Sensor Type

6 Infrared light sensor and infrared LED Line width: 0.25in minimum; optimal line width is 0.5in Response Frequency: 50Hz Light Source: GaAs infrared LED with a peak wavelength of 940nm Receiver: Si phototransistor with a sensing wavelength of 850nm (max) The range for the Line Tracker is approximately 0.02 to 0.25in (from the ground) with optimum sensitivity at 3 mm (about 1/8 inch). The minimum line width it can detect is 0.25in. Input/Output Port Assignments Line Tracker Figure 2. Three sets of three VEX line follower sensors mounted together illustrating how sensors work with examples of different readings Inputs Type: Light reflected from dark lines and a bright floor Outputs 3-Wire Cable Connect to three analog inputs Black: ground Red: +5V White: control signal The Line Tracker is an analog sensor, meaning that it can output many more values within its range of potential values (in this case, from 0-5V) than a digital sensor, which would output only a handful of discrete values in the range (e.g., 1, 2, 3, 4, and 5V), as is the case for a digital sensor. This range of output from 0-5V is sent to the microcontroller, which translates it into a corresponding range of integer values from 0 to full scale. Full scale is 1023 for 10-bit Analogto-Digital values such as with easyc or ROBOTC for PIC, 4095 for 12-bit values such as with ROBOTC for Cortex and 255 for 8-bit values such as with MPLAB. When using the VEX

7 ARM Cortex -based Microcontroller, typical white/black/"away from everything values" will be 38/662/770 for easyc, 153/2650/3076 for ROBOTC and 9/166/192 for 8-bit values. When using the PIC Microcontroller, typical white/black/"away from everything values" will be 38/882/1012 for both easyc and ROBOTC and 9/220/253 for 8-bit values. For this particular sensor, the output will be low when the surface is pale or highly reflective and high when the surface is dark and absorbs infrared energy. Ultrasonic Rangefinder Figure 2.VEX Utrasonic Rangefinder Sensor Inputs Start signal to the ultrasonic sensor. Connect to a interrupt port. Outputs 3-Wire Cable Connect to a interrupt port Echo response from the ultrasonic sensor. Black: Ground Red: +5V Orange/Yellow: Control Signal System

8 Figure 3. The developed line tracking robot. Line trackers are mounted to the back of the robot. Ultrasonic range finder is mounted on the front of the robot. iphone streams video to a computer. Lessons Learned Students are currently working on the project. Students think that this is a very interesting and challenging project. They report that they have never done something like this and are very excited about how their robot turned out and also how the robot interacted with them as a team. The streaming portion of the project was one of the highlights for the students. They used this skill for other project such as our Black History Program to stream live image of our guests and speakers from Washington D.C. Outreach activities

9 Our robotics team participates in the district UIL robotics competition with High Schools. Being part of a robotics class, instigate students to follow the STEM career pathway. Students used the streaming skills learned in this lesson for other projects such as our Black History Program to stream live speeches of our guests from Washington D.C. who spoke about the importance of being college and career oriented started in middle school. Lessons Learned The research topic sparked the interest in students making them willing to work extra hard to achieve the high expectations set by the teacher in this class. However, six weeks do not allow enough time to conduct a project of this magnitude to middle school students. The pacing works for designing the robot, building it, and programming the sensors. While working on the object recognition portion of the problem, students struggled with programming due to its complexity. The recommendation is that the robot perception part of the project be a second layer of instruction incorporating two extra six weeks into the curriculum. The first six weeks would be dedicated to research and the second six weeks dedicated to development and testing of software. Summary Students programmed the VEX sensors, which passed tests at first trial. Sensors provided robot mobility according to programmed goals. The integration of sensors as an aid to capture images and video using a cell phone camera mounted on a robot was successfully tested by students. Students were also successfully able to stream video using the iphone 6 and android cell phones. The object recognition research engaged students in learning new skills. When learning how to program using XCode and using OpenCV library students faced a greater challenge and need a greater support from the teacher. Learning how to design, build, and program robots will provide students with advanced technical professional building skills and may motivate them to pursue careers in the STEM field. Future Directions Camera programming and video processing require SDK, a system that manages the build process in an operating system, and supporting library integration knowledge. This requires computer science skills and a timeline that is beyond the scope of middle school curriculum. The development of the vision portion of robot perception research should be implemented at senior high school or college level. This process requires more than 6 weeks of research work in a K-12 classroom scenario. Acknowledgements This material is based upon work supported by the Research Experiences for Teachers Program under National Science Foundation Grant No Any opinions, finding, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of National Science Foundation.

10 Bibliography 1. Dunbar, Brian. "Machine Vision for Robotics." NASA. NASA, 09 Mar Retreived on 28 June 2015 from 2. Robot with Line Follower Sensors- Vex.YouTube. YouTube, n.d. Retrieved on 30 June 2015 from 3. ROBOTC.net. Home of the Best Robot Programming Language for Educational Robotics. Made for NXT Programming and VEX Programming. Retreived 29 June 2015 from 4. VEX Robotics. Retrieved on 29 June 2015 from