Despite the rapid progress in machine learning, sensor technology, and communication infrastructure, in certain situations self-driving cars will need human situational assessment. For example, upon recognising an obstacle on the road a request might be routed to a teleoperator, who can assess and manage the situation with the help of a dedicated workspace. Besides providing adequate views to assess the remote traffic situation, the workspace needs to enable the operator to remotely move the vehicle. A common solution to this problem is direct remote steering. However, constraints of real-world traffic scenarios, in particular the availability of high-bandwidth mobile networks, have led to concepts not relying on ultra-low latency, such as path-planning or maneuver selection. Future work, should focus on integrating different remote operating concepts into a teleoperator workspace design in order to support the variety and complexity of real-world autonomous driving challenges.