Maneuvering Experiment of Personal Mobility Vehicle with CVT-Type Steering Mechanism

Transcription

1 F2012-E Maneuvering Experiment of Personal Mobility Vehicle with CVT-Type Steering Mechanism 1 Suda, Yoshihiro * ; 1 Hirayama, Yuki; 1 Aki, Masahiko; 2 Takagi, Takafumi; 1 Institute of Industrial Science, The University of Tokyo, Japan; 2 JTEKT Corporation, Japan; KEYWORDS Personal Mobility Vehicle, CVT, Steering System, Experiment ABSTRACT This paper deals with maneuvering experiment of a tricycle-type personal mobility vehicle (PMV) with a new steering mechanism. The proposed steering mechanism has a continuously variable transmission (CVT) between rear two wheels. This feature of the PMV is steered by difference wheel speed between right and left wheels. In this paper, first the PMV mechanism is explained and then, steering characteristics is explained. Next, improved steering mechanism which makes easy to maneuver the PMV is suggested and the mechanism is explained. Finally, maneuvering experiment which was chosen as the slalom course was performed for a verification of the proposed mechanism effect. The result of the experiment shows that, due to adjusting left-side and right-side steering lever comparatively, steering characteristic has been improved. The data that has been used in this experiment are distance between tricycle and pole, usage time of riding, range of front wheel steering angle, and excessive number of steering times. INTRODUCTION In recent years, a new transportation mode which is friendly to both environment and human is required as a solution for an aging society and an environmental conservation. A personal mobility vehicle (PMV) is focused by the recent social requirement. The PMV is an individual-sized small vehicle for a short-distance and a convenient move 1-6. In our laboratory, a new concept PMV which has a combined characteristic of a bicycle and a parallel wheeled type vehicle was proposed. This PMV is able to be chosen of two modes by intended uses. The bicycle mode is appropriate for high speed riding in wide area. On the other hands, the parallel wheeled mode is expected to be used for low speed riding in crowded area. At the moment, various PMVs which have characteristic functions have been developed. In this paper, a new steering type PMV is proposed. This PMV consists of one front wheel and two rear wheels. Its steering mechanism is different from the conventional tricycle (Fig1). By using the CVT, difference of the rotation speed of the rear wheels makes the CVT steer. The objective of this paper is to propose a new human machine interface (HMI) for steering operation. Firstly the steering characteristic of this PMV is mentioned, next the new steering interface method is proposed. Finally the verify experiment method and result are shown. PROPOSED STEERING CONCEPT AND MECHANISM The PMV proposed in this paper has a different steering system from a conventional tricycle. The PMV is able to be steered by a different speed between rear two wheels which is controlled using a CVT. An overview of a PMV mechanism is shown in Figure.1. A CVT

2 output shaft is connected to a rear left wheel. A driving torque from pedals is transmitted to a rear right wheel and a CVT converted rear left wheel in travelling direction. The right wheel speed is treated as a reference wheel speed. When the left wheel speed increases by the CVT, the PMV turns to the right side. On the other hand, when the left wheel speed decreases by the CVT, the PMV turns to the left side. A driving torque difference by the CVT generates by a rotation angle of pulley which is operated by steering levers. The steering levers are connected through a wire to the pulley. A front wheel has a steering angle by a cornering force. Figure 1: Tricycle with New Control Mechanism. Figure 2: Continuously Variable Transmission Figure 3: Steering Lever and Braking Lever Figure 4: Constant Force Spring at Front Wheel

3 CHARACTERISTICS OF LEFT AND RIGHT TURINING RADIUS This steering mechanism is explained in Figure.5. When the right wheel which is the reference rotates same speed, a turning radius is different between the right turn and the left turn. Assume that the friction force between front wheel and the ground is small and vehicle attitude is determined by the difference in the rotational speed of the rear wheels, let CVT gear ratio is a, the distance between two rear wheels is r, and turning radius is R. Then, following equations are obtained. Figure 5: Mechanism of Driving Force Transmission Assume that the friction force between front wheel and the ground is small and vehicle attitude is determined by the difference in the rotational speed of the rear wheels, let CVT gear ratio is a, the distance between two rear wheels is r, and turning radius is R. Then, following equations are obtained. (1) Moreover, if we give the speed of the vehicle is V and right rare wheel speed is v, then, following equations are obtained. (2) From the equation (1), (2) this steering mechanism has the difference between turning left and turning right s turning radius and velocity.

4 Figure 6: Difference of Turning Radius between Right and Left Direction To improve the degradation of operability due to differences in turning radius of turning left and right, the author pay attention on adjusting reaction force of the left and right steering lever. In order to adjust reaction force, reaction force adjusting device was installed at the wire which connecting the steering lever and CVT input pulley. The device composes of reaction force adjusting spring inside as shown in Fig.4. Let s give the natural length is l, and the reaction force adjusting device total length is A+B, when A is the length of the screw which can be changed and B is constant value. A displacement generated from the lever action denotes x, and its stiffness denotes k m. Then a lever reaction force F m is obtained from equation (3). Then, an actual lever reaction force F L is expressed from sum of the reaction force adjusting device and a reaction force in the lever and a friction force of wire. In this paper, the length A is adjusted in the equation (4) which is adjustable as a studied parameter, in order to improve the steering operation. In the experiment, three reaction force conditions are set by adjusting of length A, shown in Table1. (3) (4) Figure 7: Internal Mechanism of Reaction Force Adjustment Device RUNNING EXPERIMENT - OVERVIEW OF EXPERIMENTAL PROCEDURE -

5 In the experiment the reaction force of left-right steering lever are changed into 3 conditions. Let s give the left hand side steering lever is Fl and the right hand side is Fr, condition1, 2, 3 is Fl=Fr, Fl>Fr, Fl<Fr respectively (Table1). The riding course which was used in the experiment is slalom course which composed of 4 poles. The distance between the poles is 7.5m, the width of course is equal to 5m and the length is 37.5m. The forward tract starts from 0 point in Fig.5, then goes the right side of the 1st pole, then passes 1st and 2nd pole to left side, then again to the right side until the end of the course respectively. The backward tract is similar to the forward tract but the tract begins from passing the 5th pole through to the left as seen in Fig.5. In the experiment, 4 male examinees are years old. The distance between the tricycle and the pole was measured by the video camera, at the same time the amount of operating lever and steering angle were measured by installed potential meters. Table 1: Conditions of Reaction Force in Experiment Condition 1 Condition 2 Condition 3 Left Reaction Force = Right Reaction Force Left Reaction Force >Right Reaction Force Left Reaction Force <Right Reaction Force Table 2: Conditions of Test Course Length [m] Course (Traveling Direction) 37.5 Interval (Traveling Direction) 7.5 Width (Cross Direction) 2.5 Figure 8: Test Course

6 RUNNING EXPERIMENT - EXPERIMENTAL RESULT - Time and distance between each pole and vehicle is measured by the camera. Let s give the name to each pole, 1-4 for forward route, 5-8 for backward route, as shown in Figure 8. Then we will get the result of the distance between each pole and vehicle as shown in Figure 9, which is the average result of 4 examinees. Plus value in the Fig.6 means the vehicle is at the right side of traveling direction and oppositely minus value means the vehicle is at the left side. For condition1 which the left and right lever s reaction is equal, when the examinees passed the poles to the right hand side, the distance between pole and vehicle is bigger than when they passed to the left hand side. The average of the different distance is 0.31m. Next comparing with condition1, condition2 which left lever reaction force is bigger than right lever, the different distance become bigger, and the average is 0.55m From the Figure 9 (b), when examinees passed to the right hand side of the pole the distance of the pole and vehicle become wider. Moreover during experiment, many contacting between pole and vehicle were noticed. Finally condition3 which left lever reaction force is adjusted to be smaller than right lever, the different distance become smaller, and the average is 0.1m. In addition, this is the only condition, which the distance between the pole and vehicle when vehicle passes to the right hand side is smaller than the left hand side. (a)condition 1 (b)condition 2 (c)condition 3 Figure 9: Average Distance between Tricycle and Pole

7 The reason why, in condition1 the distance between pole and vehicle when the vehicle passes to right hand side is bigger, is considered to be because the examines have to turn right to avoid the pole and then turn left. Because the velocity and turning radius is comparatively bigger, this results the amount of movement before turning left become bigger. Due to the reason mentioned above, in condition1 characteristic between turning left and right become different. Next when compare the condition2 with the conditon1, passing the pole to the right hand side taking more distance. On the other hands, passing the pole to the left hand side, the distance becomes smaller. This means the difference between right and left turning characteristics becomes more noticeable. Lastly, in condition3, the difference between right and left turning characteristics becomes smaller. In other words, condition3 makes the characteristic difference between left and right turning is improved. Figure 10 shows the running time in slalom course of 4 examines. From the figure, in condition3, the examines trended to finish slalom course in the shortest period of time. Although the running time is vary widely due to individual differences, running time of condition3 is the shortest compare to the two other condition. This is probably because the ease of steering results the running time to be shorter. Figure 11 shows the amplitude of the front wheel steering angle (sum of maximum angle when turning left and right). We can say that the bigger values mean the examinees made bigger turning radius. In other words, the amplitude should become smaller if the examinees ran smoothly. Therefore from Fig.8 condition3 is considered to be the smoothest running condition. In contrast, condition2 has the biggest amplitude of steering angle. This is considered to be because of the unbalance of left-right steering characteristics. Figure 10: Usage Time of Riding Figure 11: Range of Front Wheel Steering Angle

8 Figure 12: Excessive Number of Steering Times Figure 12 shows the number of excess steering times compare to the necessary times. While this number can be affected by the familiarity of the examinees, the number of excess steering time can be reduced by adjusting the left-right reaction force. Therefore from Fig.9, condition3 which has the smallest number is the finest condition to control the vehicle, compare to the other two conditions. As the conclusion, from the experiment result, it is able to reduce the different of the left-right unbalance turning characteristics by adjusting the ratio of left-right steering lever reaction force. This contributes significantly to the improvement of proposed new steering method. CONCLUSIONS In this paper, new steering method three-wheeled PMV was examined, and the conclusion is mentioned below. 1. The prototype of new steering method three wheeled PMV was made. 2. By using CVT to control the rotation speed of the left-right wheel, the PMV is possible to turn easily. 3. The relationship between the steering characteristics and operability is investigated. In a future work, this new steering type PMV is developed in order to make the further improvement. REFERENCES [1] Segway Inc.: [2] Toyota winglet: [3] Kazuo Yamafuji, Yasushi Miyagawa, Takashi Kawamura, Synchronous Steering Control of a Parallel Bicycle, Transaction of JSME, Series C, Vol.55, No.513, (1989), pp (in Japanese) [4] Hirose, N., Sukigara, K., Kajima, H., and Yamaoka, M., Mode switching control for a personal mobility robot based on initial value compensation, Proceedings of The 36th Annual Conference of the IEEE Industrial Electronics Society (IECON-2010), pp , (2010) [5] HONDA U3-X : [6] Sasaki, M., Yanagihara, N., Matsumoto, O., Komoriya, K., Steering control of the personal riding-type wheeled mobile platform (PMP), Proceedings of the 2005 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2005), pp , (2005)