Model Based Traffic Light Control For Reducing CO2 Emissions In Vehicles

Model Based Traffic Light Control For Reducing CO2 Emissions In Vehicles

S. Chitradevia, .K. Gayathrib, A. Anbarasic, C. Karthikad Guide Name: M. Gayathri Devi., M.E..,

a,b,c,dDepartment of Electronics and Communication Engineering

Anna University of Technology, chennai

Academic Campus, karudayampalayam, karur-641047.INDIA.

.

Abstract: Generally, vehicles stop-and-go driving will consume more fuel and emit more co2 than constant speed of vehicles .To reduce vehicles CO2 emissions vehicles travel should be smoothed by reducing the stop-and-go times. In this paper, three techniques are proposed to realize dynamic traffic light control for smoothing vehicles travel. In technique-1, an intelligent transportation system (ITS) employed for collecting road traffic flow data and calculating the recommended speed. In technique-2, radio antennas are installed near the traffic lights. Traffic flow information can be obtained by wireless communications between the antennas and ITC devices. In technique-3,a branch- and-bound based real time traffic light control algorithm is modeled to smooth vehicles travels. After smoothing vehicles trip, more vehicles can pass intersections with less wait time and reducing short- times stops; therefore, CO2 emissions in the vehicles can be reduced. Simulation result proves that the proposed scheme performs much better than the adaptive fuzzy light control method: The average waiting time, start-and-stop times, and CO2 emissions are efficiently reduced, and the nonstop passing rate is greatly improved.

Key words: Branch and bound (BB), Emissions of CO2, Intelligent transportation system (ITS), Three-technique open model, speed control.

1. INTRODUCTION

   Nowadays, the effect of global warming has seriously become a worldwide concern. It involves issues such as the rising of sea level problem and snow slides. During the past 100 years, the sea level has raised between10 and 25cm [1]. Our environment has changed through heat waves or extreme weather [2], [3]. To minimize the speed of our living environments deterioration, the emissions of CO2 reduction has become urgent. As one of the main sources of CO2, vehicle emissions have become more serious due to frequent increase in the number of vehicles in the worldwide [4]. A typical trip consists of idling and driving periods. The proportions of the trip spent in these stages will be based on the drivers behavior, the road type and the traffic error control. The emissions of CO2 rates are different in these stages. The two parts of the trips are 1) idling and

   2) driving. The driving period consists of cruising,

   accelerating, and deceleration [5]. Barth et al. proposed that[6], during an idling period, the engine would consume more fuel and emissions of CO2 is more than in a cruising period; on the other hand, less waiting times and a constant-speed driving style would lead to less CO2. A study in [7] indicated that CO2 emissions rate is high in decelerating than accelerating.Furthermore [6], implied that the major reason for engine idling is a stop- and-go driving style, because in a lower time period, drivers have to accelerate to go and then decelerate to stop. During this period, vehicles CO2 emissions are more. E.g., taxies waiting in queues for pick up the passengers.

   One of the important methods for smooth travel is the traffic light control model. The commonly used control scheme is the fixed-time control [8], but it cannot produce the required results. Adaptive traffic lights depend on wireless communication with the vehicles employ greater flexibility than other methods.

   Fig . .1. Three-technique open traffic light control

   model

   As above mentioned, traffic light control plays very major role in smoothing travels by minimizing vehicles waiting time and the number of stoppages[9]. Oda et al. carried out experiments a microscopic simulation model to obtain the traffic flow to reduce the waiting time in [10] and to calculate the CO2 reductions in [11]. However, the result was unclear, and it is not

   suitable for real-time applications. Alsabaan et al. had carried out by using new protocol for reducing fuel consumption and emissions in [12].

   With the development of ITS devices, it is widely used for highways by nonstop passing of vehicles [13]. In previous paper, we employed Electronic toll collection vehicles to communicate with traffic lights to obtain real-time traffic flow information and tree based algorithm was proposed to minimize the waiting time.

2. SYSTEM MODEL

   In this paper, a three-technique open model is proposed for adaptive traffic light control, which is mainly concerned at smoothing vehicles travels. In this model, we suppose that all vehicles have installed the Liquid Crystal Display (LCD) devices and ITS devices called on-board units (OBUs).LCD devices are used to collect the information, such as the speed of the current, the distance, deceleration, acceleration and the moving directions. Then, in the traffic control center, traffic lights cycles are adjusted based on the received detected traffic flow signal by a certain traffic light control algorithm. Last, an existing emission of CO2 is used to calculate the CO2 amounts

   Three-Technique Open Traffic Light Control Model:

   1. Introduction of the Three-Technique Open Model: From a macroscopic view, a three-technique open model for adaptive traffic control is shown in figure. We suppose the antennas are installed near the traffic light to collect the traffic flow information and the antenna has its own maximum distance, which is based on its performance

      1. Technique-1: The main function of technique -1 is mainly concerns collecting traffic flow information data, receiving traffic light phase data, and calculating the recommended speed. In this technique, vehicle state information will be obtained from LCD devices. The ITS OBUs devices can communicate with the traffic lights to send the current information to the traffic control center and receive the data. Based on the recommended speed, the drivers can modify or keep the current speed to pass through the intersections with less waiting time or less number of start-and-stop times.

         Fig . 2. ITS based traffic flow detection process

      2. Technique-2:In Technique-2 is mainly responsible for receiving and saving traffic flow data and then sending the control results to the ITS OBUs. This technique consists of following three parts 1) antennas,

   2. storage, and 3) traffic lights.

      As aforementioned in technique-1, the ITS OBU devices and antennas can communicate with each other by wireless communications: therefore, the real time traffic flow information can be sent to the lights. Meanwhile, the traffic control results will be sent to the ITS OBUs, and then, drivers can know the traffic light phases in time. The purpose of the storage is to save the traffic flow data that received. The traffic lights are the displays that show these control results.

      c) Technique-3: Technique-3 takes responsibility for data processing which can be divided into three parts.

      The first part is extraction of data. As vehicles regularly send the traffic flow information to the antenna before passing through the junction, it may cause some problems. For example, the same data may be received many times and the recent data may exist in the storage. Therefore it is necessary to extract the required data from the received data.

      The second part is traffic light control. The traffic control center periodically sends the ltest traffic flow data from the storage and calculates the accurate light-changing policy for this period. The optimal light- changing policy will updates new traffic flow information acquired; therefore, this shows that it is a dynamic control process. In other words, the control center will control the duration time of red or green lights phase according to the fluctuation of traffic flow. The optimal light-changing policy would be estimated by a control algorithm that can realize the lowest waiting time. Since ours is an open system others traffic light control algorithms can be used in this section, such as dynamic-programming- based fuzzy control in [14] and the decision making- algorithm in [15]; all these algorithms are acceptable for out model. Working with new control algorithms according to the intersections road situations is also available. In the following section we will provide a branch-and- bound (BB) based control algorithm.

      The third part provides an open interface for third-party applications. Vehicles information can be shared by third-party applications. For example vehicles manufacturers or magazines may use these data are used to analyze which type of vehicle used and popular brands. Moreover, real-time road traffic data can be drawn into a dynamic traffic flow map on the control center. Based on this map, the citys traffic monitor center can easily manage the traffic. In addition, the CO2 reductions can be used to enhance the quality of citys environment.

      2) Traffic flow detection process: As aforementioned, we suppose that all the vehicles have installed the ITS devices. In the following paragraphs, we will introduce the traffic flow detection process between

      the vehicles and antenna through wireless communication.

      To easily understand this process, we use three stages to describe it.

      First, when a vehicle reaches an intersection, they send the road traffic flow data to the antenna. If the vehicle comes under the coverage area of antenna, the OBU devices can directly communicate with the antenna. However if the vehicle is not present under this area, it has to send the data to the front vehicles, which can switch over the data to the antenna by multihop communication. Obviously, each vehicle performs the following two functions: 1) to relay other vehicles states data and 2) to send its own data. The OBUs periodically broadcast the vehicles motion and vehicles approach the radio coverage area, the antenna receive these states data; thus the road traffic flow information can be acquired. After data extraction, the traffic control center in technique-3 periodically picks up the traffic flow information data from the storage in technique-2. In this process, the control center acquires the recent road traffic flow before vehicles arrive at the stop line; in other words, the control center can predict the arriving vehicles queues.

      Second, after the traffic control center estimates the optimal light duration time for the next light cycle, the antenna sends the control results to the vehicles. After receiving the information from the light, OBUs will calculate a recommended speed for drivers. Then the drivers can alter to recommended speed or maintain current speed to pass through the intersections. The purpose of calculating the recommended speed is mainly concerned on the following two aspects: 1) it is to inform the drivers to how they can pass through the junctions with less waiting time and with less number of stops, and 2) it is aimed to avoid entering into the confused zone by changing the speed for the drivers safety.

      Third, after the vehicles have passed through the intersection, they also necessary to send motion states data to the antenna so that the data can be shared by the following intersections. This approach provides useful traffic flow data around the city. In short, this traffic flow detection process allow ITS devices that want to pass the intersections to send their requests to the traffic light , and then, the light adaptively changes its cycle and time duration according to the requests. But, when vehicles send traffic information to the traffic lights, the signal interfaces among vehicles would appear. Although in technique-3, the data extraction section can solve this problem in a certain degree; this problem cannot ease to zero. Therefore, how we can get a more accurate traffic information through the ITS vehicles and how we can extract the useful information from a mass of received traffic flow data still needs to be done in our future work.

      A vehicles CO2 emissions are greatly affected by the road and traffic flow conditions. The fuel consumption rates and CO2 emission rates are based on the type of road conditions. There are two types of roads rural and urban roads. Urban road usually have higher emission rates than the rural roads [6]. Oguchi et al. [16] designed an emission model for estimating vehicles CO2

      emissions using real time road traffic factors. The model is represented by the following formulas:

      E	CO2 emissions [g];
      K c	Coefficient between gasoline consume and CO2 emissions;
      D	Travel distance (in meters);
      T	Total travel time for the distance D (in seconds);
      Aee	Acceleration energy equivalent (in square meters per square seconds);
      v k	Speed at time k (in meters per second);
      o k	o k =1 if v k >v k-1 ; otherwise, k =0.

      E= Kc(0.3T=0.028D+0.056Aee) (1)

      Aee=k(vk2-vk-12) (2)

      The value of Aee can be calculated by the vehicle travel mode in acceleration and deceleration, as shown in (2). Equation (1) represents that, when the D is fixed, a vehicles CO2 emissions mainly related with the T and speed v. Hence, when the road conditions are fixed, vehicles CO2 emissions is limited to T and v. Even for the few vehicles under the same road condition, T and v will be different based on the different realistic traffic conditions. Because the traffic condition will affect the vehicles speed, acceleration, deceleration, the travel time will also changes.

3. CONTROL ALGORITHM FOR SMOOTHING TRAVEL

   To realize the aim of reducing vehicles CO2 emissions, we plan to smooth vehicles travels by the following two ways: 1) to control the traffic lights to let vehicles pass the intersections with less waiting time, which goals at reducing the idling period CO2 emissions; and 2) to recommended a suitable speed before entering the intersection, which is used to reduce vehicles driving period CO2 emissions.

   The algorithms used in this technique are,

   Algorithm 1: SBB algorithmLB searches. time LB(root)

   Branching (root)

   fori 2 to (n1 + n2 1) do

   Branching (min (LB (left node), LB (right node))) i i + 1

   time (min (LB(left node),LB(right node)) end for

   for j 1 to activenode do Branching (list[j])

   if LB(node) time then

   _ prune this branch

   else if(reach a leaf node)

   time (min (LB(left node),LB(right node)) if max _red time min _green then

   time time

   else if (time min _green) time min _green

   else if (time max _red) time max _red

   end if end if

   RETURN(time) end for

   Algorithm 2: SBB algorithmOrder search [47]. Branching (node)

   LB (node)

   _ use (3) and (4) CHANGESEQUENCE (i)

   _ change the node sequence at i LB (newnode)

   ifi< n1 + n2 1 then

   list[node] max(LB(left node),LB(right node))

   _ add active nodes to list activenode activenode + 1

   else if (max(LB(left node),LB(right node)) time

   list[node] max(LB(left node),LB(right node)) activenode activenode + 1

   end if

   1. Traffic Light Control Algorithm Problem Specification:

      When heavy traffic occurs, toevacuate the vehicles in detecting queues, we need to allocate the passing orders of the vehicles. The problem turns out to be an order-decision problem. Here, we adopt a BB [17] method to solve the order-decision problem.

      Fig . 3. Distance d and dilemma zone

      SBB Approach: The total evacuation phase consists of the following parts: 1) the time of all the detected vehicles to pass through the intersection 2) the time is wasted on changing the green lights between the roads.

   2. Calculating the recommended speed:

      The stop-and-go driving style commonly occurs when vehicles prepare to pass through the intersections.

      As aforementioned the traffic light will send the result packet to the vehicles. After receiving the packets, the

      following information can be obtained and then used to calculate the recommended speed.

      1. Distance d

      2. Current traffic light cycle, current light phase, and duration time of three phases in the current cycle Tg, Tr and Ty, where Ccycle= Tg+ Tr+ Ty;

      3. Remaining time of the current light phase Lg, Lr and Ly.

   Calculation as follows: t0 =d/Scurrent

   t1 =d/Smax Where,

   t0= time that vehicle reaches intersection

   t1=time that the vehicle pass the intersection d=distance between vehicle and next intersection

   Smax=maximum speed to cover the distance without stopping

   Scurrent=current speed of vehicle Case 1: current light phase is green:

   1. If Lg is very long that is t0 Lg, then Srec=Scurrent .

   2. If t1Lg<t0, then Srec=Smax .

   3. If t1 >Lg, then

   SRec = min (max (d/tg,Smin),Smax) . In the equation, tg stands for

   Tg=(Ng-1)CCycle + Lg + Tr + Ty + M D Where,

   Ng = (t1 Lg)/CCycle;

   D = transmission delay from signal to vehicles; M = duration time for vehicles from Scurrent to

   SRec .

   Case 2: Current Light Phase Is Red: The three cases are

   1. If Lr< to <Lr + Tg, then SRec = Scurren t.

   2. If Lr< t1 <Tg + Lr< to, then SRec = Sm+ax .

   3. If t1 >Tg + Lr + Ccycle , then SRec = min(max((d/tr),Smin ), Smax) .

      In the equation, tr = (NrCcycles + Lr + M D, and

      Nr=(t1 Lr Tg)/Ccycle Case 3: Current Light phase Is Yellow:

      1. If Ly + Tr< to _ Ly + Tr + Tg, then SRec = Scurrent .

      2. If Ly + Tr + Tg< t1 Ly + Ccycle , then SRec = Smax .

      3. If Ly + Ccycle t1 , then

         SRec = min(max(d/ty, Smin) , Smax) .

         In the equation, ty stands for ty = NyCcycle + Tr + Ly + M D ,

         andNy = (t1 Tr Ly Tg)/Ccycle . where R1=road1, R2=road 2

         REFERENCES

         1. Global Warming Signs [Date accessed: Jul. 2011]. [Online]. Available:

            http://www.john-daly.com/ges/msl-rept.htm

         2. V. Petrascu, G. Czibula, D. Sirbu, M. Popa, and D. Curseu, Identifying the impact of global warming on population using a clustering-based approach, in Proc. IEEE AQTR, Cluj-Napoca, Romania, May 2010, pp. 1 6.

         3. J. Houghton, Global warming, Rep. Prog. Phys., vol. 68, no. 6, pp. 13431403, Jun. 2005.

         4. Climate Changes Phenomenon [Date accessed: Jul. 2011]. [Online].

            Available:

            http://www.globalissues.org/issue/178/climate- change-and- global-warming

         5. CO2 The Major Cause of Global Warming [Date accessed: Jul. 2011]. [Online]. Available:

            http://timeforchange.org/CO2-cause-of-global-

            Fig . 4. Average CO2 emissions from Source to Destination.

            Fig.5. CO2 Reduction percentage from Source to Destination

4. CONCLUSION

In this paper a real time applications of traffic light control model has been proposed. This control system was constructed from a three technique open model.technique-1and technique-2 were mainly concerned with road traffic flow detection. A recommended speed calculation was carried out in technique-1, which was used to avoid dilemma zone conditions and unnecessary short time stops. With ITS vehicles the traffic control center in technique-3 could get an accurate result in real time traffic flow information. Considering drivers behavior and toleration times, an improved BB based traffic light control algorithm was designed in technique-3. Less waiting time and least short time stops could be achieved by this proposed method. Thus less CO2 emissions could also be minimized. Compared with adaptive cruising control the proposed method performed in terms of reducing the vehicles average waiting time, number of stops and emission of CO2 whereas the nonstop passing rate was also highly improved.

warming

1. M. Barth and K. Boriboonsomsin, Traffic congestion and greenhouse gases, Access, no. 35, pp. 29, Fall 2009.

2. H. C. Frey, A. Unal, N. M. Rouphail, and J. D. Colyar, On-road mea- surement of vehicle tailpipe emissions using a portable instrument, J. Air Waste Manage. Assoc., vol. 53, no. 8, pp. 9921002, Aug. 2003.

3. A. G. Sims and K. W. Dobinso, The Sydney coordinated adaptive traffic

   (SCAT) system philosophy and benefits, IEEE Trans.Veh. Technol., vol. VT-29, no. 2, pp. 130137, May 1980.

4. D. I. Robertson and R. D. Bretherton, Optimizing networks of traffic

   signals in real timeThe SCOOT method, IEEE Trans. Veh. Technol., vol. 40, no. 1, pp. 1115, Feb. 1991.

5. T. Oda, M. Kuwahara, and S. Niikura, Traffic signal control for reducing vehicle carbon dioxide emissions on an urban road network, in Proc. 11th World Congr. ITS, Nagoya, Japan, Oct. 2004, [CD- ROM].

6. T. Oguchi, M. Katakura, and M. Taniguchi,

   Carbon dioxide emis-

   sion model in actual urban road vehicular traffic conditions, JSCE J. Infrastructure Plan. Manage., no. 695/IV-54, pp. 125136, 2002 (in Japanese).

7. M. Alsabaan, K. Naik, T. Khalifa, and A. Nayak,

   Vehicular networks for reduction of fuel consumption and CO2 emission, in Proc. 8th INDIN, Osaka, Japan, Jul. 1316, 2010, pp. 671676.

8. D. A. Hensher, Electronic toll collection, Transp. Res. A: Gen., vol. 25, no. 1, pp. 916, Jan. 1991.

9. T. H. Heung, T. K. Ho, and Y. F. Fung,

   Coordinated road-junction traffic control by dynamic programming, IEEE Trans. Intell. Transp. Syst., vol. 6, no. 3, pp. 341350, Sep. 2005.

10. M. Patel and N. Ranganathan, IDUTC: An intelligent decision-making

    system for urban traffic control applications, IEEE Trans. Veh. Technol., vol. 50, no. 3, pp. 816829, May 2001.

11. C. Li and S. Shimamoto, A real-time traffic light control scheme for

    reducing vehicles CO2 emissions, in Proc. IEEE CCNC, Las Vegas, NV,2011, pp. 855859.

12. F. Yan, M. Dridi, and A. El Moudni, Control of traffic lights in intersec-

tions: A new branch-and-bound approach, in Proc. ICSSSM, Melbourne, Australia, 2008, pp. 16.