Cooperative Control of Dual-Arm Robot of Adaptive Impedance Controller Based on RBF Neural Network

Traditional impedance control methods often fail to accurately track force signals in unknown or changing environments, resulting in failure or instability in tasks such as coordinated handling. To solve the above problems, this paper adds the RBF neural network strategy, based on the traditional impedance control. First a model for the contact force between the dual redundant robotic arms and the environment is constructed, and the RBF neural network is used to estimate the stiffness of the changing environment online. Then, a dynamic adaptive force control co-simulation model is established. The change in the contact force is adapted to adjust the parameters of the two-arm impedance model to compensate for unknown environmental changes. Simulation experiments showed that the enhanced impedance control strategy is appropriate for the force-interaction circumstances of the robotic arm within the positional environment, has a stronger durability, enhances the sturdiness of the two-armed working together robot in an environment that shifts, and has a more effective force control effect.