Object Manipulation with an Anthropomorphic Robotic Hand via Deep Reinforcement Learning with a Synergy Space of Natural Hand Poses

A Reinforcement Learning Based Dirt-Exploration for Cleaning-Auditing Robot

Cleaning is one of the fundamental tasks with prime importance given in our day-to-day life. Moreover, the importance of cleaning drives the research efforts towards bringing leading edge technologies, including robotics, into the cleaning domain. However, an effective method to assess the quality of cleaning is an equally important research problem to be addressed. The primary footstep towards addressing the fundamental question of “How clean is clean” is addressed using an autonomous cleaning-auditing robot that audits the cleanliness of a given area. This research work focuses on a novel reinforcement learning-based experience-driven dirt exploration strategy for a cleaning-auditing robot. The proposed approach uses proximal policy approximation (PPO) based on-policy learning method to generate waypoints and sampling decisions to explore the probable dirt accumulation regions in a given area. The policy network is trained in multiple environments with simulated dirt patterns. Experiment trials have been conducted to validate the trained policy in both simulated and real-world environments using an in-house developed cleaning audit robot called BELUGA.

Vibrational Transportation on a Platform Subjected to Sinusoidal Displacement Cycles Employing Dry Friction Control

Currently used vibrational transportation methods are usually based on asymmetries of geometric, kinematic, wave, or time types. This paper investigates the vibrational transportation of objects on a platform that is subjected to sinusoidal displacement cycles, employing periodic dynamic dry friction control. This manner of dry friction control creates an asymmetry, which is necessary to move the object. The theoretical investigation on functional capabilities and transportation regimes was carried out using a developed parametric mathematical model, and the control parameters that determine the transportation characteristics such as velocity and direction were defined. To test the functional capabilities of the proposed method, an experimental setup was developed, and experiments were carried out. The results of the presented research indicate that the proposed method ensures smooth control of the transportation velocity in a wide range and allows it to change the direction of motion. Moreover, the proposed method offers other new functional capabilities, such as a capability to move individual objects on the same platform in opposite directions and at different velocities at the same time by imposing different friction control parameters on different regions of the platform or on different objects. In addition, objects can be subjected to translation and rotation at the same time by imposing different friction control parameters on different regions of the platform. The presented research extends the classical theory of vibrational transportation and has a practical value for industries that operate manufacturing systems performing tasks such as handling and transportation, positioning, feeding, sorting, aligning, or assembling.