Signal Processing - January 2017 - 36

analog filters to condition the sensing
signals, and digital filters implemented
by embedded software for linear, non16.5 16.5 11.5 10 10 10 10 10 10 10 11.5 16.5 16.5
linear, or adaptive signal processing.
The main source of interference on
board comes from the motor driver circuit with high peak currents up to
20 A. The noise sparks travel through
the power supply line into the control
Figure 12. The layout of the photoelectric sensor placement.
circuitry, causing erratic behaviors.
Most of the noise can be reduced by
using
analog
filters
in
the
sensing and power circuits. However,
than the nominal Nyquist frequency. Based on these sampled
some noise is still left in the sensing signal and is sampled into
data, the amplitude of the signals can be determined.
the MCU. This remaining noise can be dealt with by digital
Another example of sampling is to detect the center of
signal processing.
the racetrack. In the photoelectric sensing competition, the
Other noise sources may also be seen in each specific comsensors sample the light intensity of the racetrack surface
petition setup. For example, we previously
spatially at several isolate spots. As the
mentioned that in the two-wheel competinumber of the sensors is confined to no
Two-car chasing is
tion, a model car must keep an upright pose
more than 16 by the competition rules, the
the most challenging
while running along the racetrack. An inerarrangement of the sensors must be well
among all the competition tial measurement unit (IMU) is mounted on
designed to increase sampling range and
categories.
the chassis to measure the inclination of the
accuracy. Figure 12 shows a common laymodel car. The IMU outputs two signals:
out of photoelectric sensors in the competithe gyroscope signal that gives the angular speed, and the
tion, whereby the sensors are arranged in one line in front of
accelerometer signal that provides information on the "down"
the car with different spacing between them, and the space in
direction. Although the dip angle of the car can be calculated
the middle is smaller than the outer ones. This kind of nonaccording to the accelerometer signal alone, the movement of
uniform sampling was shown to perform better than the unithe car produces much noise mixed with the angular position.
form spatial sampling in the competition.
To address this problem, student teams have employed Kalman filters when computing the angular position (shown in
Denoising and parameter estimation
Figure 13), combining the gyroscope and accelerometer sigAn important task of the competition is to take the sensing
nals in a denoised fashion. Some student teams have simplified
signals that are often noisy or distorted and process them to
the filtering algorithm to allow for more efficient estimation of
extract important parameters or other useful information and
the attitude angle of a two-wheel race car. In addition, one of
pass it on to the control and decision mechanism to race the
the teams in the two-wheel competition carefully studied the
model car.
noise levels in the IMU data when the pulse-width modulation
Student teams have employed a variety of filters in the
(PWM) voltage with different duty ratio is applied to drive the
model car control systems to deal with the interference coming
motors. They incorporated this into a Kalman filter implemenfrom external sources or onboard circuits. They have used both
tation to further reduce the noise when estimating the pose of
the model car.

Image processing and computer vision

1
2
3

Figure 13. Using a Kalman filter to determine the angular position of a car.
Curve 1: accelerometer signal; curve 2: gyroscope signal; and curve 3: the
Kalman filter output signal.

Thanks to the power of visual sensing, student teams prefer
using cameras to guide their model cars in the competition
categories of camera sensing, chasing, and beacon-based racing. By using a camera to continuously capture images of the
racetrack, a wide range of information about the car's current
state and potential future situation can be inferred.
Detection range is a key factor affecting the achievable
speed of a self-navigating race car. Because there exists a
constant delay when a servo changes the direction of the front
wheels of a race car, detection in advance can compensate this
delay. The sooner the detection of the curve on the racetrack
is, the higher speed the race car can run without going out of
the bounds of the racetrack. Image processing from camera
data also enables the extraction of more detailed information

IEEE Signal Processing Magazine

January 2017

Table of Contents for the Digital Edition of Signal Processing - January 2017