Robotic and control systems are ever more difficult to design, program, as well as to operate. This is due to their growing size and complexity. Therefore, their real-time programming and execution architectures require more user and design assistance, based on usable abstraction and tool support. Especially in safety-critical control systems (e.g., space robotics, robotic surgery) formal methods can be useful, but they have to be made usable by application domain specialists. User-friendly, domain-specific languages and interfaces ease the application of real-time analysis and implementation techniques. We are particularly interested in task-level programming of robot and control systems, where tasks encapsulate control laws. We consider the application of discrete control synthesis techniques, integrated into the framework of a task-level robot programming environment: they support the automated generation of correct controllers for control applications. We use the synchronous programming environment Signal, and the verification and synthesis tool Sigali. This requires to determine patterns of tasks and objectives, which are at once domain-specific to control systems, and generic enough to cover a broad class of control systems. We consider its application to safe discrete teleoperation. We illustrate this by a case study concerning the interactive discrete control of tasks of an excavating system.