What is Space Robotics?
Robotics is a branch of engineering that involves the conception, design, manufacture, and operation of robots. This field overlaps with electronics, computer science, artificial intelligence, mechatronics, nanotechnology and bioengineering. Robots are machines that can be used to do jobs, some robots can do work by themselves and other robots must always have a person telling them what to do.
There are many uses robots in space. Spacecraft that explore other worlds, like the moon or Mars, are robots and are included in the “Planetary Robotics”. These include orbiters, landers and rovers to explore other planets and asteroids. The other example is “Orbital robotics”, this track support the orbital servicing and maintenance activities. Examples of these activities are the space station robotic arm that helps to build the station. The robotic arms have added new parts to the space station and move astronauts around on spacewalks. Also, the space station’s arm can move to different parts of the station, it moves along the outside of the station like an inchworm, attached at one end at a time.
Current planetary missions involve robotics vehicle to achieve the exploration goals and/or to perform the scientific experiments targeted by the corresponding exploration missions. Even though these are already very challenging and ambitious missions from a technological point of view, the global scientific objectives of these missions remain relatively modest compared to the expectations of the scientific community: exploration areas are relatively easily accessible sites, local scientific goals are pretty close to the landing site, the daily distances travelled remain quite small (typically a few tens of meters), the duration of the mission rarely exceeds 6 months, …
It can be expected that future planetary missions will aim at going a step further in any of these domains and will therefore need to have available more mature systems to help these missions gain autonomy and performances in terms of scientific return.
The following scheme represents a typical rover planetary missions aimed to reach the area of interest, perform in-situ scientific analysis, collect samples and store them into a bio-container, and transfer them between planetary assets, ready to be returned back to Earth.
In the following, the space robotics capabilities required are described in more details (referred to cyan box in the above scheme).
Vision based Autonomous Navigation (Map building & merging and self-localization)
In the context of planetary exploration this capability is of key importance due to rare communication links and is even more essential for missions like MSR where a crucial aspect is the accuracy of the localisation. In the case of a vision-based navigation, the surface vehicle relies on data fusion from its on board sensor set, which include its cameras/stereo-benches: the extraction of some visual features from the acquired images can be exploited to increase the accuracy in the vehicle relative position / orientation determination (with respect to its landing position or its operational goals) and/or its absolute position on the planet. The 3D environment reconstruction is an added output the on board visual sensor bench can provide, for benefit of the opportunistic science and on board autonomy devoted to smartly and efficiently schedule the on-the surface vehicle activities
Autonomy & Locomotion in harsh environment
Future missions foresee very complex and articulated on ground activities, potentially executed in demanding scenarios like shadowed craters, highly sloped and irregular terrain low gravity bodies for advanced exploration.
Moreover, selection of the most convenient area to perform science/detailed exploration/modules and plant settlement/ sample collection can only be accomplished after reconnaissance on a vast area. Exploring quickly and efficiently large surface areas may also require for vehicles better tuned for visiting large areas than 2D surface robots. Different surface vehicle types could be combined according to the exploration area scale and to the requirements on exploration velocity and area revisiting.
Those scenarios ask for capabilities which can be categorized into:
Relative Autonomous Navigation w.r.t. other assets (base/lander, rovers) RF and vision based
In the reference scenarios (planetary sample return) a surface robot needs to determine its own relative position with respect to other planetary assets, like landers, bases or other vehicles. This capability is crucial in exploration scenarios, which include activities to be performed and coordinated among teams of cooperative vehicles. Examples of tasks that require such functionality are 1) transport of samples collected from a specific location to a landing station for in-situ analysis or for sample return to Earth. 2) transportation of a payload from a storage place to a delivery location 3) return of a scout rover, to its carrier companion.
The capability of a mobile vehicle to identify unpredictable opportunities for science, detected while travelling in the planetary surface is key to maximize scientific return of a mission. Autonomous science requires a vehicle to be able to detect interesting regions in its surroundings, to judge and score their relevance according to the scientific mission objectives and goals, and to revise its own a-priori activity plan to include visiting and sampling of the interesting area, while respecting resource constraints.
Autonomous sample / payload manipulation, transfer between planetary assets
In the case of sample return missions, a mobile vehicle must be able to grasp an object (sample container, payload), release it from its support, transport and deliver it to its destination with a high positional accuracy. This capability relies mainly on a manipulator arm equipped with vision and force sensing. Vision based positioning is also required since the various object locations cannot be precisely known in advance. This functionality is mandatory for Mars and low gravity bodies exploration and particularly useful on the Moon to reduce the workload of human operators (TransTerra scenario).
Multi robots cooperation
In the context of planetary exploration, multi-robot systems could be essential , since a single robot is not able to cover wide and/or topologically different areas and acquire all the required information. In this view the cooperation among multiple robots allows to increase the locomotion by combining different specialized vehicles and, therefore, explore various types of terrains (a wheeled rover for even terrains and a legged scout for terrains cluttered with obstacles). The multi-vehicles asset increases the system flexibility and robustness too, by distributing tasks and functions whenever complex activities must be carried out such as large infrastructure building, elements transportation, coordinated area exploration, astronauts activities support
Reconfiguration is particularly needed in the case of multi-robots cooperation. A distributed, reconfigurable decision making architecture is required to make the team of robots working harmoniously and to adapt to the chancing configuration of the tasks and the number of robot team members. Multi-agents architectures are required to cope with coordinate local global planning scheduling, reaction, perception and reconfiguration capabilities. SW reconfiguration capabilities may require HW reconfiguration capabilities , as well. Finally standardisation in both SW and HW interfaces is required as a side aspect for reconfiguration functionalities.
Telerobotics / telepresence for rover and manipulator tasks
In presence of low latency and high bandwidth communications, like it could be for a lunar mission (TransTerra / RIMRES), telerobotics/telepresence approach is suitable to operate a robotic system. In order to allow the operator to monitor / intervene at different levels in the task execution the implementation of a flexible control / communication / human interaction architecture is necessary.
Sample collection and preparation for scientific in-situ analysis
Sample collection capability is a fundamental capability of planetary sample return missions. Different devices (micro sampler, drilling/sampling for very hard soil , very deep drill) can perform sample collection, e.g. a driller fixed on the rover or tools located at the manipulator end effector . Martian and Lunar mission require sample collection up to 2 meters in depth, dealing with different types of soil. In addition, a sample preparation and distribution functionality is necessary to allow the sample be analysed by the scientific payload an board the rover.
Sample handling for bio-container storage
After being collected the sample has to be adequately stored to prevent the backward contamination, once back on Earth. The sample bio-container and the sample handling system shall be designed to preserve the sample properties as required by scientists during manipulation, interplanetary transfer and release on Earth.
Robotics represents the active part in a servicing and maintenance scenario and it delivers the enabling technologies necessary for object handling, manipulation, assembly, etc. But in order to come to sustainable in-orbit maintenance service solutions it will be necessary to pay attention also to the counterpart, e. g. the one who is supposed to be serviced, and its capabilities of maintainability. Nowadays, spacecraft especially satellites are built as a disposal product. At the end of the life time or in case of technical problems such satellites are given up together with the mission. Mission specific monolith spacecraft design is yet not open for maintenance, repair or upgrade servicing.
On the first glance sustainability on orbital applications differs from sustainability in exploration. On orbit sustainability aims at the introduction of economical space structures and applications comparable to terrestrial solutions. Use assets as long as possible, modify and upgrade them according to changing needs and requirements, avoid wasting resources, etc. In exploration sustainability focuses more on the reuse of technological solutions for a variety of different application (different in the sense of varying environmental conditions etc.). But in any case sustainability is closely connected to standardization and modularity.
A generic model to describe on-orbit servicing missions has been devised in the course of the “Satellite Servicing Building Blocks” activity. The model allows the introduction of the different servicing operations that may require space robotics.
The operations are discussed in the following, with the intention to present the related space robotics capabilities and the likelihood these will be needed in the reference period.
Rescue mission/orbit raising: this requires a dedicated or general-purpose space tug to robotically capture (in case of uncooperative target) or robotically berth/dock (otherwise) with the stranded satellite. This operation has been proposed several times in the past and even executed by a Shuttle mission. Unfortunately the economics of the operation is totally unviable and so this operation is not likely to happen in the reference period unless cost of launch becomes sensibly lower.
Planned Orbit Raising: this concept requires a general-purpose space tug to dock or robotically berth to the client satellite(s) to pull it (them) to higher orbit. It is an attractive concept for very big space infrastructure in which the payload to be delivered in orbit is bigger than the launcher capacity. So staging of the payload is necessary. The payload is divided in multiple launches, gathered on a low orbit and then by means of the tug, pushed on the operational orbit.
Commercial infrastructure envisaged for the reference period will not require such service. However the next institutional infrastructure (post ISS) aimed at Earth Lunar L2 may benefit of such service.
Inspection/Support to Deployment operations following deployment anomaly have been studied on a number of occasions. Unfortunately this operation has proven to be economically unviable and is not likely to happen in the reference period unless the cost of launch becomes sensibly lower.
Deployment/Assembly Aid: Every spacecraft undergoes the transition between the tightly packed state at launch, which allows the spacecraft to stay within the launcher fairings, to the expanded deployed state in orbit, in which all appendages are extended to their operational state. The possibility of using spacecraft-mounted robot manipulators to deploy/assemble the appendages has been studied and also tested successfully in orbit. There is also the possibility that great part of the spacecraft may be assembled in orbit by means of dedicated robot. This would allow improving the mechanical environment of critical components at launch making spacecraft lighter and more performing.
These operations have definitely non-zero probability to happen within the reference period as the obstacles to the realisation of them are not fundamental economic limitations but just maturity of the technology involved.
Repair, Refuelling and orbit maintenance operations following failure or early depletion have been proven to be economically viable (though marginally) where the failed spacecraft belongs to a constellation and a general purpose servicer can be stationed in advance in proximity of the constellation. However these applications have not received the interest of their only potential customers: space insurers. In fact commercial spacecraft are insured by operators against failures and early depletion, so costs are borne by the insurers, which to recover the losses increase premiums to the particular operator (as well as to new insurance contracts).
Planned maintenance/refuelling, performance enhancement, additional payload and mission evolution: these operations require a client spacecraft to be re-engineered to support them. Also to be economically viable, they need to target a constellation of spacecraft all requiring operations at the same time. Of all these operations, the one that appears more likely to happen in the reference period is refuelling as it is the one that requires the least level of modification of client spacecraft.
Lifetime Extension refers to the operation that allows a telecom satellite to continue its operation despite the fact it has depleted its fuel and therefore cannot remain autonomously stationary. It can be realised by different robotic means. One form studied in the past is the so-called captive carrying. An auxiliary servicing satellite is permanently attached to the client and performs station keeping. Another more speculative way foresees a servicing satellite that periodically attaches to the client satellite and recovers the orbital drift that has occurred since its last visit. This last form can only be advantageous if there is a constellation of closely placed satellites that can be visited with negligible delta-V. Lifetime extension may be financially viable only if the whole cost of the servicing satellite is well below the revenue that the additional lifetime of client satellite(s) generate.
Even in the most benign analysis Lifetime Extension offered limited benefits and many risks, so it has failed to attract support of its possible customers (satellite operators). It is not likely that in the reference period this situation will change.
Re/De-orbiting; in recent times there has been an increased awareness of the risk posed by debris (dead satellites, upper stages and their fragments) to the operation of functional satellites. Self-disposal of satellites at the end of their operational life has become a formal practice (and in some countries obligation) for new satellites.
For satellites in GEO, re-orbiting to the Graveyard Orbit (GYO) is the required practice. In the past the use of a dedicated servicer to re-orbit constellations of telecom satellites (a sort of satellite undertaker) has been studied and found to be marginally financially viable. However the concept has failed to attract support from the potential users as GYO re-orbiting was (and is) not yet mandatory, so the percentage of operators that actually re-orbit their spent satellite is still low.
For satellites in LEO the fact that there are infinite LEO orbits makes the concept of a common servicer impossible. So new LEO spacecraft must implement independent passive or active de-orbiting means.
For all defunct satellites already in orbit, which constitute some form of threat to the usability of their orbit, the only available solution is to perform robotically Active Debris Removal (ADR).
If the threat to highly critical orbits (such as the Sun Synchronous Orbit –SSO) is confirmed, by analysis and observation of the evolution of debris population, there will be a development of ADR in the reference period.