Conventional feedback control methods can solve various types of robot control problems very efficiently by capturing the structure with explicit models, such as rigid body equations of motion. However, many control problems in modern manufacturing deal with contacts and friction, which are difficult to capture with first-order physical modeling. Hence, applying control design methodologies to these kinds of problems often results in brittle and inaccurate controllers, which have to be manually tuned for deployment. Reinforcement learning (RL) methods have been demonstrated to be capable of learning continuous robot controllers from interactions with the environment, even for problems that include friction and contacts. In this paper, we study how we can solve difficult control problems in the real world by decomposing them into a part that is solved efficiently by conventional feedback control methods, and the residual which is solved with RL. The final control policy is a superposition of both control signals. We demonstrate our approach by training an agent to successfully perform a real-world block assembly task involving contacts and unstable objects.
The mechanical behavior of OFHC copper at strain rates from 10−3 to 103 sec−1 at 300°, 420°, and 590°K was investigated. The strain rate behavior of copper can be divided into two regions. Below 10 sec−1 the dislocation motion is thermally activated over forest dislocation barriers. Above 103 sec−1, a linear relationship between stress and strain rate was observed indicating the presence of a viscous damping mechanism. The stress level τB that must be exceeded in order to obtain viscous behavior depends on the work-hardened state of the copper. The mobile dislocation density in the viscous damping region was found to be (1) independent of strain rate, (b) only a small fraction (10−5) of the total dislocation density, (c) independent of strain, and (d) increased with increasing temperature. These deductions are discussed in terms of the dislocation multiplication and annihilation mechanism.
Impact shmr tests of the Kolsky Thin Hafer type vere used. to determine tho effect of temperature and strain-rate on the critical resolved shear stress for slip in aluminu..""n sin3le crystals at strain-rates of 10 4 scc-1 and in the temperature range 20°K to 500°K. The aluminum deformed in a .vi:..>cous•rnanner in that the flow stress was proportional to the plastic strain-rate. The behavior was found to be temperature dependent. The results were discussed in terms of dislocation damping models• I•:J:-.ere friction force acting on a dislocation results :Crom, at cryogenic temperatures, electronic viscosity, and at hig:O.cr temperatures, J;honon viscosity. The theories predicted general agreement as to the ::ncc:..:;:•i tude of the observed damping but some discrepancy was found to exist bet\.recn the observed and theoretical temperature dependence of tqe d~""npin~.
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.