You are on page 1of 1

The Optimized Auto-Pilot System of Autonomous Car

Focused on Safety by Combining Ultrasonic Sensors and


Selection Sorting Algorithms
Kyohyeon Koo | kkoo4@masonlive.gmu.edu
Sittipong Ruangsang | sruangsa@masonlive.gmu.edu
Department of Computer Science & Electrical and Computer Engineering | George Mason University

INTRODUCTION AVOIDING OBSTACLES


Autopilot system
operating
Because human errors caused many car accidents (Fig. 1), autonomous
car is becoming a popular option for mobile transportation. In the field of
autopilot system, there are several kinds of autonomous car systems that During the system is
combine sensor technology with the algorithm. In this project, we are operating
proposing a system, that combines ultrasonic sensors with sorting
algorithm, for a unique self-driving car that focuses on safety.
This combination has been tested in robotics area [1], however it has not
been implemented in autopilot field. Since accidents that are caused by Obstacles detected Find and change to
autonomous cars are increasing rapidly, our project is focusing on by ultrasonic sensors another lane or way
improving the safety and solving problems that occurred in the previous
research. Our proposed system is using a combination of Seeed
Ultrasonic sensor, that has the longest sensing range, and the selection
sorting, that has a light and fast mechanism [2][4]. Computing by
selection sorting
algorithm Avoid obstacles

Figure 3. How Sorting Algorithm Works

Sorting algorithm (Fig. 3) is simpler and faster than other algorithms [2]. Still Obstacle
So, this algorithm is best option for operating autopilot system. Selection detected
sorting use measured data based on the ultrasonic sensors. First step is
making a new heap sort from the raw data. Second step is to process the Figure 4. System Flowchart
Figure 1. Percentage of Car Accidents caused by human error [8] data and provide proper commands to cars quickly and accurately [1].

Conclusion
METHODOLOGY COMBINING ULTRASONIC SENSORS AND SELECTION
DETECTING OBSTACLES SORTING ALGORITHMS The primary purpose of this proposal is to determine the most optimal
auto-pilot system that utilizes ultrasonic sensor technology and selection
Based on the previous experiments, it demonstrates that the combination
sorting algorithms for safety driving. We hope the high performance of the
of ultrasonic sensor and sorting algorithm on mobile robot, gives
auto-pilot system in the autonomous car will change the car accident by
capabilities for the robot to move around, detect and avoid obstacles
almost 0 percent. Our suggestion for future research is to optimize the
efficiently [5]. Moreover, it can precisely measure the distance of its
autopilot to make it react faster. Moreover, the optimization can prevent
surroundings and its exact position and orientation in its workspace [6].
leaking memories and control the usage of CPU memory reasonably.
The experiments that are conducted on mobile robot, have fully
demonstrated the possibility of implementing the ultrasonic sensor and
sorting algorithm on autonomous car [7]. So, we're trying to incorporate
these researchers' work on robots into self-driving systems.
References
[1] RadarMAC: Mitigating Radar Interference in Self-Driving Cars, 2016, IEEE. Joud Khoury, Ram Ramanathan, Dan
Figure 2. How Ultrasonic Sensor Works McCloskey, Russell Smith, Timothy Campbell

An ultrasonic sensor (Fig. 2) is a device that sends electromagnetic HOW THE PROPOSED SYSTEM WORKS [2] Md. Azharul Islam and Shawkh Ibne Rashid, “Algorithm for Ethical Decision Making at Times of Accidents for
Autonomous Vehicles”, IEEE Conference 2018
waves to measure the distance of objects. To avoid accidents or crashes, Fig. 4 depicts our proposed system in a flowchart. The system begins with [3] A. Iqbal, S. Ahmed, M. Tauqeer, A. Sultan, and S. Abbas,
the autonomous car should be able to detect obstacles to operate the autopilot start operating, then the ultrasonic sensors send “Design of multifunctional autonomous car using ultrasonic and infrared sensors”, IEEE Conference Publication.
smoothly and efficiently. The distance of the obstacles from the electromagnetic waves to detect and measure the distance of obstacles. [4] Seeed Ultrasonic Sensor. (2019). www.robotshop.com.
autonomous cars should also be measured. It calculates the distances of After that, the autopilot system use selection sorting algorithm to avoid the [5] F. Aulian and J. Agung “Mapping algorithm using ultrasonic and compass sensor on autonomous mobile robot”
objects within a radius of 200 cm in real time [3]. In our project, we will obstacles. The autonomous car changes to another lane and repeats the [6] C. Beakauk “Mobile robot position control algorithm based on multiple ultrasonic distance sensors”

improve the range of obstacle detection by using the Seeed Ultrasonic same process until the car stops. [7] Geometry Based Approach to Obstacle Avoidance of Triomnidirectional Wheeled Mobile Robotic Platform

Sensors which can measure up to 400 cm of distance[4]. [8] Autonomous Cars Infographic group work.(2019) www.infograph.venngage.com

You might also like