The robotic platform in this study has dual, seven-function, hydraulically actuated manipulators, which are being used for research into assisted tele-operation for common nuclear decommissioning tasks, such as pipe cutting. The article concerns the identification of state-dependent parameter (SDP) models for joint angle control. Compared to earlier SDP analysis of the same device, the present work proposes a new way of representing the state-dependent gain and parametrises this using novel regret-regression methods. A mechanistic interpretation of this model yields dead-zone and angular velocity saturation coefficients, and facilitates SDP-based control with an inverse dead-zone. This approach integrates the input signal calibration, system identification and nonlinear control system design steps, using a relatively small data-set, allowing for straightforward recalibration when the dynamic characteristics have changed due to age and use, or after the installation of replacement parts.