Machine Learning-Based Cognitive Position and Force Controls for Power-Assisted Human–Robot Collaborative Manipulation

Manipulation of heavy objects in industries is very necessary, but manual manipulation is tedious, adversely affects a worker’s health and safety, and reduces efficiency. On the contrary, autonomous robots are not flexible to manipulate heavy objects. Hence, we proposed human–robot systems, such as power assist systems, to manipulate heavy objects in industries. Again, the selection of appropriate control methods as well as inclusion of human factors in the controls is important to make the systems human friendly. However, existing power assist systems do not address these issues properly. Hence, we present a 1-DoF (degree of freedom) testbed power assist robotic system for lifting different objects. We also included a human factor, such as weight perception (a cognitive cue), in the robotic system dynamics and derived several position and force control strategies/methods for the system based on the human-centric dynamics. We developed a reinforcement learning method to predict the control parameters producing the best/optimal control performance. We also derived a novel adaptive control algorithm based on human characteristics. We experimentally evaluated those control methods and compared the system performance between the control methods. Results showed that both position and force controls produced satisfactory performance, but the position control produced significantly better performance than the force controls. We then proposed using the results to design control methods for power assist robotic systems for handling large and heavy materials and objects in various industries, which may improve human–robot interactions (HRIs) and system performance.

Granular scaling laws for helically driven dynamics

Exploration of granular physics for three-dimensional geometries interacting with deformable media is crucial for further understanding of granular mechanics and vehicle-terrain dynamics. A modular screw propelled vehicle is, therefore, designed for testing the accuracy of a novel helical granular scaling law in predicting vehicle translational velocity and power. A dimensional analysis is performed on the vehicle and screw pontoons. Two additional pontoon pairs of increased size and mass are determined from dimensional scalars. The power and velocity of these larger pairs are predicted by the smaller pair using the scaling relationships. All three sets are subjected to ten trials of five angular velocities ranging from 13.7 to 75.0 revolutions per minute in a high interlock lunar regolith analog derived from mining tailings. Experimental agreement for prediction of power (3-9% error) and translational velocity (2-12% error) are observed. A similar set of geometries is subjected to multibody dynamics and discrete element method cosimulations of Earth and lunar gravity to verify a gravity-dependent subset of the scaling laws. These simulations show agreement (under 5% error for all sets) and support law validity for gravity between Earth and lunar magnitude. These results support further expansion of granular scaling models to enable prediction for vehicle-terrain dynamics for a variety of environments and geometries.

The Robot Selection Problem for Mini-Parallel Kinematic Machines: A Task-Driven Approach to the Selection Attributes Identification

In the last decades, the Robot Selection Problem (RSP) has been widely investigated, and the importance of properly structuring the decision problem has been stated. Crucial aspect in this process is the correct identification of the robot attributes, which should be limited in number as much as possible, but should be also able to detect at best the peculiar requirements of specific applications. Literature describes several attributes examples, but mainly dedicated to traditional industrial tasks, and applied to the selection of conventional industrial robots. After a synthetic review of the robot attributes depicted in the RSP literature, presented with a custom taxonomy, this paper proposes a set of possible requirements for the selection problem of small scale parallel kinematic machines (PKMs). The RSP is based on a task-driven approach: two mini-manipulators are compared as equivalent linear actuators to be integrated within a more complex system, for the application in both an industrial and a biomedical environment. The set of identified criteria for the two environments is proposed in the results and investigated with respect to working conditions and context in the discussion, emphasizing limits and strength points of this approach; finally, the conclusions synthesizes the main results.