Soft robots undergo large nonlinear spatial deformations due to both inherent actuation and external loading. The physics underlying these deformations is complex, and often requires intricate analytical and numerical models. The complexity of these models may render traditional modelbased control difficult and unsuitable. Model-free methods offer an alternative for analyzing the behavior of such complex systems without the need for elaborate modeling techniques. In this paper, we present a model-free approach for open loop position control of a soft spatial continuum arm, based on deep reinforcement learning. The continuum arm is pneumatically actuated and attains a spatial workspace by a combination of unidirectional bending and bidirectional torsional deformation. We use Deep-Q Learning with experience replay to train the system in simulation. The efficacy and robustness of the control policy obtained from the system is validated both in simulation and on the continuum arm prototype for varying external loading conditions.
There are a number of instances in nature where long and slender objects are grasped by a continuum arm spirally twirling around the object, thereby increasing the area of contact and stability between the gripper and the object. This paper investigates the design and modeling of spiral grippers using pneumatic fiber-reinforced actuators. The paper proposes two reduced order models, a pure helical model, and a spatial Cosserat rod model to capture the deformed behavior of the gripper using the mechanics of fiber-reinforced actuators in the presence of self-weight. While the former model can yield closed form expressions that aid in design, the deformation parameters deviate by greater than 40% of its length for actuators longer than 200 mm. However, the Cosserat rod model deviates by less than 8% of its length for two different prototypes validated in this work. The deformation of the gripper is then correlated to the number of spiral turns achievable about the object, which determines the quality of the grip. Together, they enable a systematic framework where the gripper parameters can be designed for a given range of object sizes to be handled. This framework is experimentally validated by successful gripping of a range of slender objects that lie between a 20 mm diameter tubelight and a 60 mm diameter PVC pipe.
Soft robots undergo large nonlinear spatial deformations due to both inherent actuation and external loading. The physics underlying these deformations is complex, and often requires intricate analytical and numerical models. The complexity of these models may render traditional model based control difficult and unsuitable. Model-free methods offer an alternative for analyzing the behavior of such complex systems without the need for elaborate modeling techniques.In this paper, we present a model-free approach for open loop position control of a soft spatial continuum arm, based on deep reinforcement learning. The continuum arm is pneumatically actuated and attains a spatial workspace by a combination ofunidirectional bending and bidirectional torsional deformation. We use Deep-Q Learning with experience replay to train the system in simulation. The efficacy and robustness of the control policy obtained from the system is validated both in simulation and on the continuum arm prototype for varying external loading conditions
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.