This paper presents a methodology for planning the orientation of the end-effector for an industrial robot based on the application of augmented reality. The targeted applications are those where the end-effector is constrained to follow a visible path, which position and model are unknown, at suitable inclination angles with respect to the path. The proposed approach enables the users to create a list of control points interactively on a parameterized curve model, define the orientation of the end-effector associated with each control point, and generate a ruled surface representing the path to be planned. An approximated time-optimal trajectory, which is a determined subject to robot actuators and joint velocity constraints using convex optimization techniques, is implemented to simulate a virtual robot, allowing the users to visually evaluate the trajectory planning process. A case study is presented and discussed.