Before entering into additional details of the ERGO system, it is important to recall what is autonomy for a robotic system. For the scope of ERGO the definition of on-board autonomy as from the European ECSS Space Segment Operability Standard is assumed:
“On-board autonomy management addresses all aspects of on-board autonomous functions that provide the space segment with the capability to continue mission operations and to survive critical situations without relying on ground segment intervention”.
These ECSS autonomy levels are summarized in Table 1‑2. Although system autonomy is often assimilated to on-board autonomy, it is clear that also the ground segment plays a crucial role in any space system. Therefore, if performances improvements at system level want to be assessed for a certain system, a sufficient level of autonomy shall be included also in the ground segment part, aimed to support the operations of an autonomous spacecraft system.
|E1||Mission execution from ground control; limited onboard capability for safety issues||Real-time control from ground for nominal operations. Execution of time-tagged commands for safety issues||Real-time control with pre-programmed sequences|
|E2||Execution of pre-planned, ground-defined, mission operations on-board||Capability to store time-based commands in an on-board scheduler||Pre-planned|
|E3||Execution of adaptive mission operations on-board||Event-based autonomous operations. Execution of on-board operations control procedures||Semi-autonomous, also called "Adaptive"|
|E4||Execution of goal-oriented mission operations on-board||Goal-oriented mission (re-)planning||Goal-Oriented Operation|
Table 1‑2 Autonomy Levels as defined in the European ECSS Space Segment Operability Standard
From an architectural point of view, the consequences of increasing the autonomy of a ground system, a spacecraft or a rover, imply changes in both the control architecture and the commanding protocol: due to the fact that most of the environmental conditions are not known in advance, it is impossible to predict the behaviour, and the system behaves as a non-deterministic, since it is not precisely known the sequence of low-level commands that will be executed when a high-level command is issued. Autonomous robotic explorers must be proactive in the pursuit of goals as well as reactive to evolving conditions in the environment. These concerns must be balanced over short and long time horizons, considering timeliness, safety, and efficiency, presenting a substantial challenge for control system design. In addition, new approaches for verification and validation are needed
The difficulty to achieve/pursue autonomy strongly depends on the assumptions to be considered. Table 1‑3 identifies the most relevant factors applicable to the problems of interest that are key to properly define the requirements of the autonomous system.
|Environment with high uncertainty||Three sources of uncertainty: Partial observability which leads to a partial understanding of the state of the world; Non-determinism appears in real-world domains because actions can lead to different possible states; A dynamic domain could spontaneously change its state due to external events.|
|Limited On-board Resources||Available on-board resources, especially in terms of computing power, memory and energy, are limited.|
|Limited Communications||Communications to the robotic mean might be limited due to obstacles (e.g. operations inside a crater) , long delays (e.g. interplanetary missions) or communication windows|
|Highly complex operations||Increasing payload and platform capabilities enable the achievement of more complex missions.|
|Criticality||Spacecraft represent critical systems for which high safety standards must be enforced.|
Table 1‑3: Factors that influence the level of autonomy in application domains (orbital and surface)