서강대학교

초록/요약

The thesis aims to propose mechanism design guidelines, apply the proper mechanism design to a robotic leg of legged robots, and implement running. Running requires fast motion during the swing, great force production during the stance, and impact mitigation at the transition from swing to stance to legs. Therefore, running is believed to be the most challenging task that the legs must deal with. Leg mechanisms that can be divided into actuator configuration, segment ratio, overall length, and flexion direction are hypothesized to affect fast swing recovery, great force production, and impact mitigation, as mentioned above. The study has proved the hypothesis and further, design a robotic leg with running-adept mechanisms to implement running. The first objective is to define the metrics that can measure the important features of running, i.e., fast swing recovery, great force production, and impact mitigation, and analyze the mechanisms, i.e., actuator configuration, segment ratio, overall length, and flexion direction, with the metrics. The metrics of tangential mobility and radial force producibility are defined corresponding to the running requirements of fast swing recovery and great force production while the metric of maximum contact inertia is defined corresponding to the running requirement of impact mitigation. Each mechanism is intensively analyzed in view of these three metrics, i.e., tangential mobility, radial force producibility, and maximum contact inertia. Experimental results verify that the three metrics are affected by the mechanisms. The second objective is to propose mechanism design guidelines for running based on the previous analysis. Since tangential mobility and radial force producibility represent the tangential velocity of the tip with respect to joint angular velocities and the radial force of the tip with respect to joint torques, respectively, the larger the better. On the other hand, since maximum contact inertia represents the effective inertia felt at the tip, the smaller the better. According to the previous analysis, the mechanism of overall length has a clear trade-off. For example, the long overall length leads to the large tangential mobility and maximum contact inertia, and the small radial force production. Conversely, the short overall length leads to the large radial force production and maximum contact inertia, and the small tangential mobility. Therefore, the overall length is excluded from the design guidelines and instead, the overall length is fixed as a constant. Since actuator configuration is divided into serial and parallel depending on how the actuators are mounted while flexion direction is divided into backward and forward depending on the leg's heading direction, both of them have two discrete variables, which leads to two discrete values in terms of tangential mobility, radial force producibility, and maximum contact inertia, respectively. On the other hand, since segment ratio has infinite number of ratio as the length of upper and lower limbs changes, it has continuous variables, which leads to continuous values in terms of tangential mobility, radial force producibility, and maximum contact inertia, respectively. Comprehensively considering feasibility conditions and practical constraints, the mechanism design guidelines are proposed in the end. According to the design guidelines, a leg is designed, fabricated, and implemented for running. Depending on the application, the mechanisms can be chosen differently. For example, one might decide to select the backward flexion, even if the decision compromises the effect of impact mitigation, in order to secure the sight of the robot. In either way, the metrics of tangential mobility, radial force producibility, and maximum contact inertia are found to be very useful in that they vary on the mechanism design related to actuator configuration, segment ratio, and flexion direction.