for Automatic Parking
Sharath Panduraj Baliga,Automotive
software engineering,TU –Chemnitz ,Chemnitz,Germany.
Email: [email protected]
Abstract— The automactic parking system is one among
most researched topics to help the drivers to park their vehicle with ease even
in the narrow paths and make the parking management systems simple for the
increasing number of vehicles in the major cities.It is very helpful for the
drivers if there is some system which informs them about the available parking
space in the parking area.This paper provides the walk through to the various
available algorithms for automatic parking system. The ideas behind every
algorithm which concentrate on development of different stages of the automatic
parking are discussed.
Keywords: AVM-Around View Monitor, LSD – Line segment detector,
RGB – Red Green and Blue.
is a very difficult task even for a skilled driver because of limited space,
incoming vehicles, and fixed and moving obstacles such as pedestrians.
Therefore, the development of parking assistant systems and autonomous parking
systems is important. Furthermore, an autonomous vehicle must perform numerous
tasks to safely park the vehicle in a narrow space, including precise environment
detection and parking manoeuvres 5,18.
Automatic parking also helps in reduction of traffic problems, since the 20% of
all the congestion in the major cities are caused by frustrated drivers driving
around the block searching for parking spaces 14.
Many automatic parking systems have been proposed
maximize both the safety and convenience of
parking. Target position designation is one of the primary components of an automatic
parking system. Perception methods to detect
available parking spaces are categorized into two
space-based approaches 2-5
and parking slot marking-based approaches 6-15
rest of this paper is organised as follows: Section-2 gives the important areas
of concentration for automatic parking, Section-3 will brief about state of
art, Section-4 discusses about the various algorithm and the concepts, Section-5
concludes the paper and Section-6 gives the future scope.
2 Important Steps
in Automatic Parking:
the free parking plot.
the steering input and velocity of the vehicle (Including accelerating or
braking to track the desired trajectory). 5,18
3 State of art:
commercial version of automatic parallel parking was introduced by Toyota Motor
Corporation in Toyota Prius in 2004. Lexus also debuted a car, the 2007 LS,
with an Advanced Parking Guidance System17, while the detection of parking
slot was done using ultrasonic sensors and now it’s been evolved and replaced
by vision based detection.
parking system has evolved from semi-automatic to automatic, up to the level
where the parking can be done with the help of remote control or the mobile app
in latest BMW cars.4
Parking Employing Swarm Algorithm.
Generation of shortest possible path
trajectory for the parallel parking with one shot track according to the dimensions
of the car.
A fuzzy controller is designed to decide
the driving velocity and the required steering angle.
The fish swarm search algorithm is used to
search for fuzzy controller parameters and the parking time that achieve best
path-planning algorithm for parallel automatic parking.
The method overcomes the following
drawbacks of previous researches
The path generation using clothoid curve
where it’s difficult to approximate holonomic path by smooth non-holonomic path
which may cause more operations in the parking process.8,13
Motion generation with trigonometric
function that plans a continuous and iterative path. But the parking space
should be larger than other method to avoid moving forward and backward too
This method gives an iterative
path-planning algorithm to park in the narrow space which is not big enough to
operate parking in one time.19
4.3 A Trajectory Planning Method Based on Forward
Path Generation and Backward Tracking Algorithm.
Assuming the free parking plot is detected,
this algorithm plans for generation of parking path from a desired starting
point and the heading angle to the destination with referred heading angle in
the parking lot.15
The algorithm plans the backward trajectory by a forward path heading
out from parking to the desired position on the road.9
The forward path is implemented by dividing the path in to two segments
(locating segment and the entering segment) .9
The arc in the locating segment and the Bezier curve in the entering
segment are connected together. The arc can consider the minimum distance between
the front corner of the vehicle and a static obstacle using the geometric
This algorithm can be applied to both parallel parking and the
4.4 A Benchmark and a Learning-Based Approach
for Visual based Parking slot detection
A large scale parking slot image database
is established and for each image in the database marking points and parking
slots are carefully labelled, this database is used as benchmark to detect the
While in learning based approach, for the
given test image the marking points will be detected first and then valid
parking slots can be inferred. 10
4.5 Parking Slot Detection Based on Around View
Monitor (AVM) Systems.
Detecting the parking slot with the help of
vision based approach i.e. around view monitor offers 360 degrees surrounding
vehicle view by summing up all the images obtained by 4 fisheye cameras located
at centre of front bumpers, rear trunk lid and each of the side B-pillar.16
This method utilizes line segment detector
(LSD) to detect parking slot marking lines which have parallel line pair and
this method of detection is found to be faster. It also overcomes the drawback
of most of the vision based approach for parking slot detection while detecting
the parking spot where the markings are damaged.2,3
Location of cameras in AVM system.
parking space detection.
This system helps the drivers to
have an idea about the free parking space in the parking yard so that the time
in searching for the free space will be reduced.
The algorithm for
the same is as follows:
System will get Livestream
video of the parking lot from camera.
Images are captured when a car
enters or leaves the parking lot.
RGB Images are converted to
Select the coordinates of the parking lot.This
will crop the extra space other than parking lot from the image.
Select the coordinates of the single parking
slot. This will divide the parking lot into equal size slots.
Each block is converted from
grayscale to binary and then inverse binary to get the car in white color and parking
area into black color.
Threshold value is calculated
in every block to detect whether that block contain car or not.
If value is less than threshold
value than that block is free and available for parking car and if value is
greater than block is occupied.
There are different algorithms to realize
automatic parking system. The parking slot detection can be done by ultrasonic
or visual based methods, but the visual based method overcomes some of the
drawbacks like dependencies where the method using ultrasonic sensors requires
the reference object behind the vehicle to fetch the path.
Most of the parking
algorithm uses fuzzy logic control to tract the path towards the parking slot.
Each algorithm looks like a better approach overcoming
the drawbacks of the past researches. Hence research under automatic parking
system still exists and works towards the development of more smooth and
reliable parking system.
In future this paper work can be taken as benchmark
and convert in to a research work on the specific area in automatic parking
Research forums like IEEE explorer and
Google Scholar are the main sources for the research papers referred in this
1 Bae, II, et
al. “Path Generation and Tracking Based on a Bezier Curve for a Steering
Rate Controller of Autonomous Vehilcles.” Intelligent Transportation
Wang, H. Zhang, M. Yang, X. Wang, L. Ye, and C. Guo, “Automatic
based on a bird’s eye view vision system,” Adv. Mech. Eng., vol.
pp. 1–13, Dec. 2014.
2Gioi R G,
Jakubowicz J, Morel J M, et al. “LSD: a line segment detector”. Image
Processing On Line, vol. 2, no. 4, pp. 35-55, 2012.
3Gioi R G, Jakubowicz
J, Morel J M, et al. “LSD: A fast line segment detector with a false detection
control”. IEEE transactions on pattern analysis and machine intelligence,vol.
32, no. 4, pp. 722-732, 2010.
Jung, D. S. Kim, P. J. Yoon, and J. H. Kim, “3D vision system for
recognition of free parking site location,” Int. J. Autom. Technol., vol.
no. 3, pp. 361–367, May 2006
5Idris, M. Y.
1., et al. “Car Park System: A Review of Smart Parking System and its
Technology.” Information Technology Journal 8.2,2009.
Paromtchik and C. Laugier, “Autonomous parallel parking of a nonholonomic
vehicle”, in Proc. IEEE Intelligent Vehicles Symp.,Tokyo, Japan, Sep. 1999, pp.
Paromtchik, and C. Laugier, “Motion Generation and Control for Parking an
Autonomous Vehicle,” Proc. of the IEEE Int. Conj. On Robotics and Automation,
Minneapolis, USA, April 22-28, 1996, pp.3117-3122.
Laumond, P.E. Jacobs, M.Taix, R.M.Murray, “A motion planner for nonholonomic
mobile robots”, IEEE Trans. Robot. Autom., 1994,10, (5), pp. 577–593
Moon, II Bae, Jae-gwang Cha and Shiho Kim “A Trajectory Planning Method
Based on Forward Path Generation and Backward Tracking Algorithm for Automatic
Parking Systems” 2014 Intelligent Transportation Systems (ITSC) October
8-11,2014. Qingdao, China
H.G., Kim, D.S., Yoon, P.J., Kim, J., “Parking slot markings
for automatic parking assist system,” IEEE Intelligent
Symposium, pp. 106-113, June 2006.
Lin Zhang1, Xiyuan Li, Xiao Liu, Ying Shen and Lu Xiong “Vision-Based
Parking-Slot Detection: A Benchmark And A Learning-Based Approach” IEEE International
Conference on Multimedia and Expo (ICME) 2017
Li, Changle Li, Qieshi Zhang,Tao Guo and Zhifang Miao “Automatic Parking Slot
Detection Based on Around View Monitor (AVM) Systems “2017 9th International
Conference on Wireless Communications and Signal Processing (WCSP)
Hana.fy*, Mostafa M Gomaa, Mohamed Taher, Ayman M Wahba .” Path Generation and
Tracking for Car Automatic Parking Employing Swarm Algorithm” The 2011
International Conference on Computer Engineering & Systems, Nov. 29
2011-Dec. 1 2011
DEUTSCHER J., GRODDE S, “Continuous curvature trajectory design and feed forward
control for parking a car”, IEEE Trans. Control Syst. Technol., 2007, 15, (3),
Nazia Bibi, Muhammad Nadeem Majid ,
Hassan Dawood, and Ping guo. “Automatic Parking Space Detection System” , 2nd International Conference on Multimedia and Image Processing,2017
Mendiratta, Debopam Dey, Deepika Rani Sona “Automatic Car Parking System with
Visual Indicator along with IoT” 2017 International conference on Microelectronic
Devices, Circuits and Systems (ICMDCS)
Hisashi, et al. “Development of parking space detection using an
ultrasonic sensor.” Proc. 13th World Congr. Intell. Transp.Syst. Serv.,
Jeong, C. G. Choi, J. N. Oh, P. J. Yoon, B. S. Kim, M. Kim, and K.
Lee, “Low cost design of parallel parking assist system based on an
sensor,” Int. J. Autom. Technol., vol. 11, no. 3, pp. 409–
16Suhr J K,
Jung H G. “Fully-automatic recognition of various parking slot markings in
Around View Monitor (AVM) image sequences”,2012.
Hsu, Jing-Fu Liu, Pen-Ning Yu, Wang-Shuan Lee and Jia-Sing Hsu .” Development
of an Automatic Parking System
IEEE Vehicle Power and Propulsion Conference (VPPC), September 3-5, 2008,
Massaki, Kang Sup Yoon, and Hideki Hashimoto.”Development of advanced
parking assistance system.” Industrial Electronics, IEEE Transactions on 5
Lv, Linhui Zhao, Zhiyuan Liu “A path-planning algorithm for parallel automatic
parking “Third International Conference on Instrumentation, Measurement,
Computer, Communication and Control ,2013.