Technical Feasibility Study on the Personal Rapid Transit (PRT) Pilot

Transcription

1 Technical Feasibility Study on the Personal Rapid Transit (PRT) Pilot System 1 Yeun-Sub Byun, 2 Rag-Gyo Jeong, 3 Seok-Won Kang, 4 Jong-Gyu Hwang, 5 Sung-Il Seo 1,First Author On-demand Transit Research Team, Korea Railroad Research Institute, ysbyun@krri.re.kr 2 On-demand Transit Research Team, Korea Railroad Research Institute, rgjeong@krri.re.kr *3,Corresponding Author On-demand Transit Research Team, Korea Railroad Research Institute, swkang@krri.re.kr 4 On-demand Transit Research Team, Korea Railroad Research Institute, jghwang@krri.re.kr 5 New Transportation Systems Research Center, Korea Railroad Research Institute, siseo@krri.re.kr Abstract This paper presents an alternative approach to the conventional personal rapid transit (PRT) concept for extended application as a complementary transportation mode. In particular, the Korean PRT system employs less visually obstructive and economic infrastructures for steering control and power transfer. The guidance control based on magnetic markers reference sensing is robust and inexpensive. Furthermore, the track designed for automatic guidance is simply constructed in a paved lane with minimal markings. In addition, Inductive Power Transfer (IPT) a supercapacitor hybrid energy supply system requires lower capital expense for construction of power feeding lines as well as harmonized connection on streets with mixed traffic. The physical and operational feasibility of these features are assessed for a pilot network implemented on the grounds of the Korea Railroad Research Institute (KRRI). Finally, this study contributes to the reduction of risk for further investment and to the identification of major requirements and prerequisites. Keywords: Inductive Power Transfer (IPT), Hybrid Energy Supply System, Magnetic Markers Reference Sensing, Personal Rapid Transit (PRT) 1. Introduction The personal rapid transit (PRT) system has emerged from efforts to complement the role of mass transportation systems as well as to offer mobility services effectively similar to a private car characterized by door-to-door and demand-responsive transport capability [1]. PRT vehicles are typically defined as a four to six passenger vehicles with fully automated driving capabilities that operate on a network of roads that offer optimized routes and immediate on-demand transport without time delays [1]. There are two systems (i.e., ULTra and 2getthere) officially operated on pilot basis in the world, and great effort is being made to meet the increasing demand for personal public transportation. In connection with this, Korea Railroad Research Institute (KRRI) in South Korea has performed an implementation study for the development of a pilot PRT system suitable as a new transportation mode with competitiveness as well as strategic novelty as a latecomer since The implementation study was based on the previous feasibility study conducted in The classical concept of a PRT system has not been successfully realized as envisioned by proponents due to the lack of a possible implementation strategy from a physical and economic feasibility perspective. Furthermore, the original proposals were too advanced for the technology available at that time (i.e., the 1970s). However, the theory remains active, and this originally unrealistic concept is becoming more feasible due to the development of electric vehicle (EV) technology. In this study, the representative technical features of the Korean PRT system [2] are reviewed with a possible implementation scenario and experimentally tested by modified golf cart to check the feasibility of prospective technologies and to identify user requirements on a pilot network constructed at KRRI. International Journal of Digital Content Technology and its Applications(JDCTA) Volume 7, Number 12, August

2 2. Implementation scenario A Korean PRT vehicle is being developed based on the fundamental architecture of small EVs, but robust autonomous driving as well as wireless power transfer capability is also being considered for extended application. The use of independent infrastructures for steering control, such as guideways and power supply by physical contact has resulted in the delay of widespread introduction of a PRT system. On the other hand, guidance control by magnetic-marker reference sensing is technically reliable and economically efficient. Also, the inductive power transfer (IPT)-supercapacitor hybrid energy supply system has the advantages of high power density and rapid charging. These two characteristics are intended to cause minimal interference with the existing transportation infrastructures and ease harmonized connection with them. 3. Guidance control system 3.1. System architecture In general, PRT is autonomously operated on the network of roads independently designated for the system, and passengers can select their destinations from a touch screen-based kiosk at the station. Once the signal for remote calling of the vehicle is delivered to an operational control center, the vehicle in the nearest possible position is offered to serve the transport services. The main role of the operational control center arranges and allocates the called vehicle to provide optimal transport scheduling. The vehicle requested by the passengers is prepared at the station platform to provide high accessibility to the passengers, and when the starting button is pressed by passengers on board, the vehicle immediately begins to run toward the destination. The initially allocated route is updated according to traffic flow conditions or changed in case of unexpected events, such as avoiding obstacles in its path or collision with preceding vehicles. Most importantly, the real-time recognition of the self-position and orientation of the vehicle on the network is essential to guide the vehicle by the optimal route to the destination without delay in time. To date, numerous efforts for tracking vehicle position in real-time have been made, and it has been realized by the technological advancement in sensing and visualization. Methods to monitor vehicle position are mainly classified by the type of referenced information; that is, absolute position determination and relative position estimation methods. The absolute position determination method includes GPS [3-4] measurements or magnetic markers reference sensing. On the other hand, relative position estimation is achieved by means of a high-precision visualization technique using cameras [4-5] or infra-red laser scanners [6] in addition to recognition by connection with a power transmission line. In this study, autonomous guidance control over a vehicle is implemented in a manner that identifies the vehicle s location by detecting the magnetic markers embedded in a paved roadway at specific intervals. Figure 1 shows the operational concept of the guidance system for a Korean PRT vehicle. Figure 1. Schematic of guidance system architecture for steering control 328

3 3.2. Magnetic sensing ruler (MSR) In the magnetic marker-based guidance system, the position estimation of a vehicle is an important task for steering the vehicle along the path to its destination. For this purpose, the open guideway is simply constructed by positioning the magnetic markers under the surface of paved road, and a geographic database of GPS logs of magnetic markers is created and recorded in the vehicle host computer. In addition, the magnetic sensing ruler (MSR) assembled by three modular boards, each with seven Hall sensor arrays, is mounted at the bumper position parallel to the axle axis as shown in Figure 2 for detecting the magnetic markers. When a vehicle is passing over the magnetic markers on the trajectory, the MSR can detect the magnetic fields emanating from the markers. The MSR is connected to the vehicle host computer by an RS232 communication link, and the detected signals from the MSR are processed in the main controller (i.e., host computer) for steering control. Figure 2. Magnetic sensing ruler (MSR) mounted on the vehicle 3.3. Steering control mechanism The test vehicle was prepared by modifying a used golf cart. which is generally the most simplified electric vehicle (EV) platform. Basically, a steering motor and the main driving motor along with its driver have been already equipped in a golf cart; therefore, in addition to the MSR, as shown in Figure 3, various sensors for detecting the steering angle, vehicle speed, and angular velocity to monitor the kinematic information of the operating vehicle and a steering motor driver for vehicle guidance control are also installed. The on-board computer, used as a main controller for steering control, is dedicated for calculation of the absolute position (x, y) and orientation (θ) of the vehicle by referring to the realtime information on the vehicle status collected from the MSR and other kinematic sensors during its driving mode. The position and orientation data is used to identify the position of a vehicle on a path. Figure 3. Schematic block diagram of autonomous guidance control system based on magneticmarkers reference sensing 329

4 The routine selected by passengers is activated by configuring the database consisting of the absolute coordinate points of each magnetic marker on the trajectory. The vehicle identifies its position by mapping the measured coordinate values onto those in the database, and then the vehicle is controlled by reference to the precursors (i.e., the next magnetic markers) pre-defined in the trajectory. In addition, this database includes permitted speed limits and information on station locations. Thus, providing the target destination and current position of the vehicle is given, the vehicle is steered in accordance with the allocated trajectory, and the operating speed of the vehicle is controlled by considering user comfort as well as the safety of the vehicle at the starting point, at intermediate stopping points, or on curved roads Software for user interface To efficiently communicate between the user and the autonomous vehicle and to achieve operational stability during driving mode, the human-machine interface (HMI) software was developed, which is operated on the host computer (i.e., main controller). When the users select their routine after boarding, the appropriate path is activated and displayed on the map (e.g., the red line shown in Figure 4). This touch screen-friendly software (programmed by Visual C++, Microsoft) enables tusers to request their transport and to display vital information on the vehicle. Real-time information regarding the vehicle status, such as its position, orientation, and steering velocity is described as shown in Figure 4 on the on-board computer. Figure 4. Screen shot of the main software running on the on-board computer for communication with passengers 4. Power supply system 4.1. Inductive power transfer system (IPTS) The supercapacitor is advantageous in terms of power density and the rate of charging, but it is inherently impossible to offer long-duration operations and the required high volume for the same power density as compared with the Li-ion battery. Thus, the IPT-supercapacitor hybrid energy system [7] is expected to complement the weakness of the supercapacitor and offer an adequate level of transport services that can be defined in the implementation scenario for the test-bed network. In addition, high-speed stationary charging at a station enables the reduction of magnetic flux leakage and construction cost. In particular, lateral displacement error (i.e., lateral misalignment between the primary coil shown in Figure 5 and pick-up coils on-board the vehicle) in vehicle control is directly associated with the efficiency of the IPTS [8]. 330

5 Figure 5. Embedded primary coil and inverter for resonance of magnetic flux 4.2. Energy storage media Initially, the Li-ion battery or supercapacitor was primarily considered as the energy storage devices as shown in Table 1. The Li-ion battery was generally preferred for most EV applications, including PRT systems (e.g., 2getthere: LiFePO4 battery). Because of its long charging time, high initial cost, and short life time, it is limited for use in standard energy storage. Because the highest priority was given to high-speed charging combined with IPT technology, the supercapacitor was considered preferable in this study. Furthermore, efforts (e.g., carbon nanotube-based supercapacitor or ultracapacitor (LIC: Li-ion-based supercapacitor)) to improve the main weakness (i.e., low energy density) of the supercapacitor are continuing. Table 1. Comparison of the characteristics of the Li-ion battery and the supercapacitor Specifications Sections Referenced products Charging time Weight Life time Li-ion battery Q (Samsung SDI) 3 hours 503 kg 500 cycle Supercapacitor EMHSR-0166C0-048R0S (Nesscap) within 2 min 60 kg 500 thousands ~ 1 million cycle 5. Testbed-based assessment The pilot network was designed to demonstrate the operational feasibility, and its total length is approximately 650 m. A total of 313 magnetic markers are embedded in the testbed route, and the interval between them is 3 m as shown in Figure 6(a). There are five shelters (as shown in Figure 7(a)) to provide real-time scheduling information for waiting and boarding of passengers and to deliver electric power to the pick-up coil on-board the vehicle from road-buried power rails. Figure 6(b) represents the experimental operation of the pilot vehicle developed in this study. Currently, the operating speed of a vehicle is controlled within 15 km/h, but it will be enhanced up to 40 km/h according to the developing schedule for stabilization of the control algorithm. In addition, the charging performance was evaluated in a stationary mode at the station, and 10 kw output power with 81% efficiency with a 10 cm air gap between power lines buried in the ground and pick-up coils onboard is achieved as depicted in Figure 7(b). 331

6 (a) Figure 6. (a) Design of pilot route for system assessment and (b) experimental demonstration of autonomous driving of a vehicle on the pilot network (b) (a) Figure 7. (a) PRT station and infrastructure for power charging by IPT method and (b) characteristics of charging voltage and power of IPTS 6. Summary and conclusion The possible implementation scenario of the Korean PRT system and testbed-driven assessment of its pilot system were presented in this study. The system performance did not completely satisfy the requirements predefined as target specifications of this study. However, the successful operation of the pilot system has demonstrated the substantial potential for the proposed approach for guidance control and power transfer. Eventually, the evaluation results will be reflected at a later stage of developments which will be carried out for further improvement. 7. Acknowledgements This work was supported by a research grant from Korea Railroad Research Institute (KRRI), titled Development of Technologies for On-demand Circulator Transit System (PRT) (grant no. PK13006A), Republic of Korea. 8. References [1] Jack H. Irving, Harry Bernstein, C.L. Olson, Jon Buyan, Fundamentals of Personal Rapid Transit, D.C. Health and Company, Lexington M.A., [2] Rag-Gyo Jeong, et al., Development of Technologies for On-demand Circulator Transit System (PRT), Korea Railroad Research Institute (KRRI) 1st year report, (b) 332

7 [3] T. Upadhyay, S. Cotterill, A.W. Deaton, Autonomous GPS/INS Navigation Experiment for Space Transfer Vehicle, IEEE transactions on Aerospace and Electronic Systems, vol. 29, no. 3, pp , [4] Byeong-Mook Jung, In-Joo Yeo, Che-Seung Cho, Autonomous Traveling of Unmanned Golf-Car using GPS and Vision System, Journal of the Korean Society for Precision Engineering, vol. 26, no. 6, pp , [5] Young-Jae Ryoo, Young-Cheol Lim, Visual Control of Autonomous Vehicle by Neural Networks using Fuzzy-Supervised Learning, Journal of Electrical Engineering and Information Science, vol. 2, no. 2, pp , [6] Jens Christian Andersen, Nils A. Andersen, Ole Ravn, Vision Assisted Laser Scanner Navigation for Autonomous Robots, Experimental Robotics. Springer Berlin Heidelberg, [7] Baek-Hyun Kim, Rag-Gyo Jeong, Sang-Gi Chung, Seok-Won Kang, A Development of the Electric Power Supply System for PRT Vehicle, The Transactions on the Korean Institute of Electrical Engineers, vol. 62, no. 2, pp , [8] Seungyoung Ahn, Yangbae Chun, Dong-Ho Cho, Joungho Kim, Wireless Power Transfer Technology in On-Line Electric Vehicle, Journal of the Korean Institute of Electromagnetic Engineering and Science, vol. 11, no. 3, pp ,