Latest Papers

ASME Journal of Mechanisms and Robotics

  • Mechanical Characterization of Supernumerary Robotic Tails for Human Balance Augmentation
    on August 31, 2023 at 12:00 am

    AbstractHumans are intrinsically unstable in quiet stance from a rigid body system viewpoint; however, they maintain balance, thanks to neuro-muscular sensory control properties. With increasing levels of balance related incidents in industrial and ageing populations globally each year, the development of assistive mechanisms to augment human balance is paramount. This work investigates the mechanical characteristics of kinematically dissimilar one and two degrees-of-freedom (DoF) supernumerary robotic tails for balance augmentation. Through dynamic simulations and manipulability assessments, the importance of variable coupling inertia in creating a sufficient reaction torque is highlighted. It is shown that two-DoF tails with solely revolute joints are best suited to address the balance augmentation issue. Within the two-DoF options, the characteristics of open versus closed loop tails are investigated, with the ultimate design selection requiring trade-offs between environmental workspace, biomechanical factors, and manufacturing ease to be made.

“On the Edge” Obstacle Surmounting Method Using Hybrid Locomotion


This paper presents on the edge obstacle surmounting method for QuadRunner, a hybrid quadruped robot, to overcome obstacles using hybrid locomotion where both legged and wheel configurations are utilized. When obstacle heights exceed the workspace of its leg, QuadRunner becomes quasi-statically mismatched, meaning the robot’s kinematic constraints are not satisfied, and it fails to achieve the climbing task quasi-statically. By incorporating its body as contact support, the center of gravity (COG) of QuadRunner can be successfully shifted on top of the obstacle to perform surmounting task. The unique design of the QuadRunner leg allows it to behave as a four-bar or slider-crank mechanism depending on the leg’s configuration. Here, we detail the sub-state strategy for its surmount task, where QuadRunner goes through the sub-states {L}EAN, {H}OOK, {F}OUR-BAR, {S}LIDE, {G}ET-UP to climb obstacles. In addition, limitations of the operation are analyzed and the requirements for climbing are identified. With our proposed method, QuadRunner can surmount obstacles of heights between 10 cm and 22 cm (higher than its kinematic max height of 16 cm) within 25 s. Lastly, a reliability test shows that the robot can climb the obstacle with a 70% success rate.

Read More

Journal of Mechanisms and Robotics Open Issues