Who is not familiar with the struggle of parking in urban areas? The car was built to provide mobility and flexibility, however with the number of cars in the world exceeding one billion (Greencarreports.com, 2014), parking struggles do not necessarily increase mobility and flexibility. The increased number of cars, and lack of parking space in urban areas results in difficulty for drivers to find a parking space. In turn this results in unnecessary road use and thus traffic congestion, increased road heat, air pollution and noise pollution (Hagman, 2006).
Horng proposes an innovative adaptive recommendation mechanism for smart parking. This mechanism will be capable of controlling vehicular traffic by recommending drivers with a parking spot, and by providing the optimal directions towards the specific parking place. The main benefits of this proposed model are reduced traffic congestion, reduced time searching for a parking spot both resulting in decreased road heat, air pollution and noise pollution.
The mechanism provides the driver with a parking recommendation and directions to this location based on the location of the driver, the parking space requirement and traffic surrounding the driver. In technical terms, vehicle to road (V2R) and vehicle to vehicle (V2V) communications take place by means of on-board units (OBUs) and road-side units (RSUs). An RF module transmits this location, the parking requirement and congestion information to the parking congestion cloud (PCCC) when a parking request is submitted.
Based on the vehicle location, nearest parking lot, parking lot status and the driving direction the PCCC computes the optimal parking solution and the optimal route to this parking place. This optimal parking solution is based on several technical concepts I will briefly explain. The Cellular Automata (CA) process converges information of single participants, cells, into a description of the system’s global behavior (Missoum, Gürdal and Setoodeh, 2005). Each driver is modelled as a cell whose state depends only on the prior state of cells in a well-defined neighborhood. The CA model is used to combine information from multiple drivers at the same time to avoid congestion. Further more the Artificial Fish Swarm Algorithm (AFSA), proposed by Li Xiao Lei in 2002, is used to model the parking behavior of drivers. AFSA imitates fish behavior such as preying, following and swarming with the local search of individual fish to reach a global optimum. The PCCC uses AFSA for the recommendation and compares parking behavior to fish behavior where searching for parking is seen as preying, and swarming is related to drivers’ behavior that is influenced by the behavior of other drivers. The most innovative technological advance in telecommunications is Cognitive Radio (CR). CR establishes ubiquitous connections which intelligently take temporal and spatial unused resources into account, and is therefor highly mobile and able to adapt itself to changing environments. In the recommendation system CR is used for stable V2R and V2V connections, necessary to keep a stable connection for cars driving at high speed. The figure below demonstrates the system of cars sending and receiving information.
The proposed model is most applicable in urban areas where parking space is limited and congestion, noise and air pollution are major problems. Horng has simulated the model in an urban area of 5km2 x 5km2. The practical application of the model will most certainly occur in larger areas. Currently many applications are implemented on navigation devices to tell vehicle location, nearby parking locations and the distance until the end destination. However these systems don’t take real-time information on other road users and congestion into account, when recommending a parking location.
For further research I would suggest researching drivers’ opinion towards such recommendation system. Drivers could be reluctant to follow the recommendation if the model suggests different parking spaces then used before, and in that case how does the system adapt to drivers not following the recommendation?
In conclusion the proposed adaptive parking recommendation mechanism decreases parking space search time, increases the probability and success rate of finding an available parking space, decreases congestion and air and noise pollution. It provides convenience and decreases frustration for the driver. Altogether the only thing you need to do is drive your car following the recommended directions, and within years the car might even drive itself there …
Horng, Gwo-Jiun. “The adaptive recommendation mechanism for distributed parking service in smart city.” Wireless Personal Communications 80.1 (2015): 395-413.
http://www.greencarreports.com/news/1093560_1-2-billion-vehicles-on-worlds-roads-now-2-billion-by-2035-report Accessed 10 February 2016
Hagman, Olle. “Morning queues and parking problems. On the broken promises of the automobile.” Mobilities 1.1 (2006): 63-74.
Missoum, S., Z. Gürdal, and S. Setoodeh. “Study of a new local update scheme for cellular automata in structural design.” Structural and multidisciplinary optimization 29.2 (2005): 103-112.