Patent Application
20230077168
The present invention relates to a method of landing an aircraft, and an aircraft comprising a landing system configured to land the aircraft.
To aid pilots in landing scenarios a brake-to-vacate system can determine a landing plan for the aircraft which will optimise a landing to vacate from the runway at an appropriate exit.
When approaching to land, there are a variety of factors that may drive how the aircraft decelerates and which runway exits it takes (i.e. the landing plan). These include: time to reach the gate, time to allow brakes to cool for the next flight, amount of brake wear, amount of tyre wear, fuel burn during landing, and taxiing to reach the gate.
A first aspect of the invention provides a method of landing an aircraft, the aircraft comprising a first aircraft, the first aircraft comprising a landing system, the method comprising operating the landing system to: receive information from a second aircraft via a direct aircraft-to-aircraft communication from the second aircraft to the first aircraft; determine a landing plan of the first aircraft based on the information; and land the first aircraft based on the landing plan of the first aircraft.
The use of a direct aircraft-to-aircraft communication (i.e. a communication from the second aircraft to the first aircraft which does not travel via an intermediary such as a land station, air traffic controller or satellite) makes the method reliable because such a communication is inherently secure and difficult to hack.
The landing plan of the first aircraft may comprise at least one of: an approach air speed, a runway speed profile, a braking command profile, a thrust reverser command profile, a spoiler deployment profile, an exit from the runway, and a taxiing route.
The landing plan of the first aircraft may be determined by revising an initial landing plan of the first aircraft.
The landing plan of the first aircraft may be determined by an optimisation algorithm, such as an evolutionary algorithm.
The first and second aircraft may be members of a virtual fleet of aircraft, and the optimisation algorithm may determine an optimised set of landing plans for the virtual fleet of aircraft which is optimised for the entire virtual fleet. The virtual fleet may have only two members (i.e. the first and second aircraft) but more typically the virtual fleet has three, four or more members.
The information received from the second aircraft may comprise information associated with the second aircraft, such as a static aircraft parameter of the second aircraft (for example weight); status information of the second aircraft (for example brake wear, tyre wear, fuel amount); schedule information of the second aircraft (for example acceptable delay or scheduled arrival time); and/or a landing plan of the second aircraft (for example an approach air speed, a runway speed profile, a braking command profile, a thrust reverser command profile, a spoiler deployment profile, an exit from the runway, a taxiing route etc.).
The landing system may be operated to receive further information associated with a third aircraft; and determine the landing plan of the first aircraft based on the information associated with the second aircraft and the information associated with the third aircraft.
The further information may be received via a direct aircraft-to-aircraft communication to the first aircraft from the second aircraft (for example as part of a “daisy-chain” network) or from the third aircraft (for example as part of a “hub and spoke” network).
The landing plan of the first aircraft may be transmitted to the second aircraft.
The landing plan of the first aircraft may be transmitted to the second aircraft via a direct aircraft-to-aircraft communication from the first aircraft to the second aircraft. A landing system of the second aircraft may determine a landing plan of the second aircraft based on the landing plan of the first aircraft.
A second aspect of the invention provides an aircraft comprising a landing system configured to land the aircraft by the method of the first aspect, wherein the aircraft is a first aircraft, the aircraft landing system comprising: a receiver arranged to receive information from a second aircraft via a direct aircraft-to-aircraft communication from the second aircraft to the first aircraft; a processor arranged to determine a landing plan of the first aircraft based on the information; and one or more actuators arranged to land the first aircraft based on the landing plan of the first aircraft.
A further aspect of the invention provides a pair of aircraft comprising: a first aircraft according to the second aspect; and a second aircraft arranged to transmit the information to the first aircraft via the direct aircraft-to-aircraft communication from the second aircraft to the first aircraft.
A further aspect of the invention provides a method of landing an aircraft, the aircraft comprising a first aircraft, wherein the first aircraft is a member of a virtual fleet of aircraft and the first aircraft comprises a landing system, the method comprising operating the landing system to: receive landing plans of all other aircraft of the virtual fleet; input the landing plans of all other aircraft of the virtual fleet into an optimisation algorithm which determines a set of landing plans for the virtual fleet of aircraft which is optimised for the entire virtual fleet, the optimised set of landing plans for the virtual fleet of aircraft including a landing plan of the first aircraft; and land the first aircraft based on the landing plan of the first aircraft.
The virtual fleet may have only two members, but more typically the virtual fleet has three, four or more members.
Embodiments of the invention will now be described with reference to the accompanying drawings, in which:
Each aircraft 1-4 has an initial landing plan. This plan may include an approach air speed, a runway speed profile, a braking command profile, a thrust reverser command profile, a spoiler deployment profile, an exit from the runway, and/or a taxiing route etc.
An initial landing plan of the first aircraft 1 is given below in Table 1.
The initial landing plan of the first aircraft 1 is graphically illustrated in
As shown in
An initial landing plan of the second aircraft 2 is given below in Table 2.
The initial landing plan of the second aircraft 2 is graphically illustrated in
As shown in
This process continues in a “daisy-chain” from the first aircraft 1 to the fourth aircraft 4. Thus the third aircraft 3 forwards the landing plans 10, 20 to the fourth aircraft 4, and also sends the fourth aircraft 4 its own landing plan 30. The daisy-chain may continue further to include many more aircraft.
A similar daisy-chain communication network operates in the opposite direction. Thus the fourth aircraft 4 sends the third aircraft 3 its landing plan 40 (optionally along with landing plans from any other aircraft behind the fourth aircraft 4) via a direct aircraft-to-aircraft communication 4c; the third aircraft 3 sends the second aircraft 2 the landing plans 30, 40 via a direct aircraft-to-aircraft communication 3c; and the second aircraft 2 sends the first aircraft 1 the landing plans 20, 30, 40 via a direct aircraft-to-aircraft communication 2c.
The two-way daisy-chain communication network of
The set of landing plans can then be analysed collectively at each aircraft by an optimisation algorithm run on-board the aircraft by a landing system of the aircraft, and changed if necessary to provide an optimised set of landing plans for the aircraft 1-4 which are optimal in a collective sense. In other words the collective landing plan (i.e. the set of landing plans) of the virtual fleet is optimised, rather than landing plans being optimised on an individual basis.
The direct aircraft-to-aircraft communications 1b, 2b, 3b, 2c, 3c, 4c avoid transmitting information to a land station, air traffic controller, satellite or other intermediary which is not a member of the virtual fleet. Such direct aircraft-to-aircraft communications are inherently secure and difficult to hack. This increases the certainty that the overall optimisation of the virtual fleet is close-to-optimal, and that the determined set of landing plans is indeed close-to-optimal. Each communication may be further secured by an encryption algorithm, or other layers of communication security.
Table 3 and
The revised landing plan 20a includes a revised runway speed profile 21a which reduces to zero at a runway distance of 2000 ft (rather than 1000 ft as in the initial runway speed profile 20) and a revised braking command profile 22a which involves lighter braking than the initial braking command profile 22. The revised landing plan 20a may result from the fact that the third aircraft 3 and/or the fourth aircraft 4 have a more pressing need to use gates associated with the early runway exit that the second aircraft 2 was planning to use as part of its initial landing plan 20. So in the revised landing plan 20a, the second aircraft 2 relaxes its braking plan to go to the next runway exit.
Table 4 and
The revised landing plan 10a includes a runway speed profile 11a which reduces to zero at a runway distance of 2000 ft (rather than 2750 ft as in the initial runway speed profile 11), and a braking command profile 12a which is heavier than the initial braking command profile 12. The revised landing plan 10a may result from the fact that the following aircraft 2-4 have a pressing need to minimise delays, so in the revised landing plan 10a the first aircraft 1 increases its braking to exit the runway earlier.
In the case of
The optimised set of landing plans 10a, 20a, 30, 40 is optimised for the entire virtual fleet 1-4, rather than for any one member of the virtual fleet.
After the first aircraft 1 lands, the second aircraft 2 becomes the lead aircraft and the process repeats as shown in
The optimisation algorithm seeks to optimise parameters for all of the aircraft in the virtual fleet. Parameters can be at least one of: total airport delay; total aircraft fuel; total aircraft brake and tyre wear; and total aircraft delay. The optimisation algorithm may assign weights to each parameter to represent the relative importance of each parameter.
Determining an optimal set of landing plans for a virtual fleet with a low number of aircraft may be a fairly simple calculation if the number of aircraft in the virtual fleet is low (in this case, four aircraft). However, if the virtual fleet has many aircraft, then running the optimisation algorithm 100 on an aircraft may become computationally challenging. This is partly due to the number of parameters which may need to be considered by the algorithm (i.e. runway speed profiles, braking command profiles, landing plans, intended runway exit, intended runway stopping distance, etc.) and the inter-relationships between different parameters and for each aircraft, as well as information about the airport (e.g. number of runways, shared taxiing routes, location/distance of runway exit on each runway, etc.).
A particularly computationally efficient form for the optimisation algorithm 100 is an evolutionary algorithm, such as a genetic algorithm or a swarm algorithm. For example, an evolutionary algorithm may generate an initial population of sets of landing plans for the virtual fleet; breed new sets of landing plans with an element of mutation; evaluate the individual fitness of the new sets; and replace the least-fit sets of landing plans with new sets of landing plans.
The evolutionary algorithm may run continuously, repeatedly modifying the population of sets of landing plans. At a threshold, either of distance to runway or estimated time to landing, the lead aircraft identifies the most-fit set of landing plans in the population (this most-fit set including the final landing plan of the lead aircraft); transmits its final landing plan to air-traffic control; and then lands at the airport based on its final landing plan. This process can avoid negotiations with air-traffic control and increase airport thoroughput and efficiency and reduce aircraft maintenance.
In the examples above, a set of landing plans of all aircraft of the virtual fleet is input into the optimisation algorithm, which then generates a set of optimised landing plans. In an alternative example, other types of information may be input into the optimisation algorithm 100 instead of a set of landing plans. For example, the inputs into the optimisation algorithm 100 may be static aircraft parameters for all aircraft of the virtual fleet (for instance the weights or types of the aircraft); status information for all aircraft of the virtual fleet (for instance brake wear, tyre wear or fuel amount); schedule information for all aircraft of the virtual fleet (for example acceptable delay or scheduled arrival time); and/or any other type of information which may be required to calculate an optimal set of landing plans.
So in this case, such other types of information may be transmitted via the direct aircraft-to-aircraft communications 1b, 2b, 3b, 2c, 3c, 4c instead of the landing plans 10, 20, 30, 40 shown in
Alternatively, the direct aircraft-to-aircraft communications 1b, 2b, 3b, 2c, 3c, 4c may include such other types of information in addition to the landing plans 10, 20, 30, 40. So such other types of information may be input into the optimisation algorithm along with the landing plans.
In the example of
Certain elements of
Each aircraft 1-4 comprises a landing system 31 as shown in
The landing plan of the aircraft may be displayed to the pilot by a display 35 and/or transmitted to a following aircraft via a transmitter 36.
One or more actuators 37 (such as brakes, thrust reversers, spoilers etc.) are arranged to land the aircraft based on its landing plan. Thus when it is time to land the aircraft, the processor 33 sends control commands to the various actuators 37 to land the aircraft based on its final landing plan. The processor 33 may be part of the aircraft's auto-flight computer, and may run “brake-to-vacate” software which controls the landing of the aircraft based on its final landing plan.
Where the word ‘or’ appears this is to be construed to mean ‘and/or’ such that items referred to are not necessarily mutually exclusive and may be used in any appropriate combination.
Although the invention has been described above with reference to one or more preferred embodiments, it will be appreciated that various changes or modifications may be made without departing from the scope of the invention as defined in the appended claims.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2001122.7 | Jan 2020 | GB | national |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/EP2021/051012 | 1/19/2021 | WO |
