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.
In order to understand infection of avian influenza A virus (AIV) and canine distemper virus (CDV) in the Siberian Tiger in Northeast China, 75 Siberian Tiger serum samples from three cap- tive facilities in northeastern China were collected. AIV and CDV antibody surveillance was test- ed by using hemagglutination inhibition and serum neutralization methods. The results showed that the seroprevalence of H5 AIV, H9 AIV and CDV was respectively 9.33% (7/75), 61.33% (46/75) and 16% (12/75). In the 1<years <2 and > 5 year-old group, the seroprevalence of the H9 AIV was 24% and 80% (P < 0.01), and the CDV seroprevalence was 6% and 36% (P < 0.01), respectively. It was demonstrated that 3 (4%) out of 75 serum samples were AIV+CDV seropos- itive, with 2.67% (2/75) in H9+AIV and 1.33% (1/75) in H5+H9+AIV. To our knowledge, this is the first report of AIV and CDV seroprevalence in Siberian Tigers in China, which will provide base-line data for the control of AIV and CDV infection in Siberian Tigers in China.