Compared with the robots, humans can learn to perform various contact tasks in unstructured environments by modulating arm impedance characteristics. In this article, we consider endowing this compliant ability to the industrial robots to effectively learn to perform repetitive force-sensitive tasks. Current learning impedance control methods usually suffer from inefficiency. This paper establishes an efficient variable impedance control method. To improve the learning efficiency, we employ the probabilistic Gaussian process model as the transition dynamics of the system for internal simulation, permitting long-term inference and planning in a Bayesian manner. Then, the optimal impedance regulation strategy is searched using a model-based reinforcement learning algorithm. The effectiveness and efficiency of the proposed method are verified through force control tasks using a 6-DoFs Reinovo industrial manipulator.
Sapelovirus A (SV-A) is a positive-sense single-stranded RNA virus which is associated with acute diarrhea, pneumonia and reproductive disorders. The virus capsid is composed of four proteins, and the functions of the structural proteins are unclear. In this study, we expressed SV-A structural protein VP1 and studied its antigenicity and immunogenicity. SDS-PAGE analysis revealed that the target gene was expressed at high levels at 0.6 mM concentration of IPTG for 24 h. The mouse polyclonal antibody against SV-A VP1 protein was produced and reached a high antiserum titer (1: 2,048,000). Immunized mice sera with the recombinant SV-A VP1 protein showed specific recognition of purified VP1 protein by western blot assay and could recognize native SV-A VP1 protein in PK-15 cells infected with SV-A by indirect immunofluorescence assay. The successfully purified recombinant protein was able to preserve its antigenic determinants and the generated mouse anti-SV-A VP1 antibodies could recognize native SV-A, which may have the potential to be used to detect SV-A infection in pigs.