| Nov. 27, 2023, 2:11 p.m. |
Added
35
|
{"external_links": []}
|
|
| Nov. 20, 2023, 2:02 p.m. |
Added
35
|
{"external_links": []}
|
|
| Nov. 13, 2023, 1:33 p.m. |
Added
35
|
{"external_links": []}
|
|
| Nov. 6, 2023, 1:30 p.m. |
Added
35
|
{"external_links": []}
|
|
| Aug. 14, 2023, 1:30 p.m. |
Added
35
|
{"external_links": []}
|
|
| Aug. 7, 2023, 1:31 p.m. |
Added
35
|
{"external_links": []}
|
|
| July 31, 2023, 1:33 p.m. |
Added
35
|
{"external_links": []}
|
|
| July 24, 2023, 1:34 p.m. |
Added
35
|
{"external_links": []}
|
|
| July 17, 2023, 1:33 p.m. |
Added
35
|
{"external_links": []}
|
|
| July 10, 2023, 1:25 p.m. |
Added
35
|
{"external_links": []}
|
|
| July 3, 2023, 1:26 p.m. |
Added
35
|
{"external_links": []}
|
|
| June 26, 2023, 1:25 p.m. |
Added
35
|
{"external_links": []}
|
|
| June 19, 2023, 1:26 p.m. |
Added
35
|
{"external_links": []}
|
|
| June 12, 2023, 1:28 p.m. |
Added
35
|
{"external_links": []}
|
|
| June 5, 2023, 1:32 p.m. |
Added
35
|
{"external_links": []}
|
|
| May 29, 2023, 1:27 p.m. |
Added
35
|
{"external_links": []}
|
|
| May 22, 2023, 1:28 p.m. |
Added
35
|
{"external_links": []}
|
|
| May 15, 2023, 1:30 p.m. |
Added
35
|
{"external_links": []}
|
|
| May 8, 2023, 1:36 p.m. |
Added
35
|
{"external_links": []}
|
|
| May 1, 2023, 1:27 p.m. |
Added
35
|
{"external_links": []}
|
|
| April 24, 2023, 1:34 p.m. |
Added
35
|
{"external_links": []}
|
|
| April 17, 2023, 1:29 p.m. |
Added
35
|
{"external_links": []}
|
|
| April 10, 2023, 1:24 p.m. |
Added
35
|
{"external_links": []}
|
|
| April 3, 2023, 1:26 p.m. |
Added
35
|
{"external_links": []}
|
|
| Jan. 28, 2023, 11:08 a.m. |
Created
43
|
[{"model": "core.projectfund", "pk": 23890, "fields": {"project": 1073, "organisation": 4, "amount": 561557, "start_date": "2019-03-31", "end_date": "2021-09-29", "raw_data": 38289}}]
|
|
| Jan. 28, 2023, 11:08 a.m. |
Created
40
|
[{"model": "core.projectperson", "pk": 53723, "fields": {"project": 1073, "person": 11587, "role": "PM_PER"}}]
|
|
| Jan. 28, 2023, 10:51 a.m. |
Added
35
|
{"external_links": []}
|
|
| April 11, 2022, 3:45 a.m. |
Created
43
|
[{"model": "core.projectfund", "pk": 15989, "fields": {"project": 1073, "organisation": 4, "amount": 561557, "start_date": "2019-03-31", "end_date": "2021-09-29", "raw_data": 6309}}]
|
|
| April 11, 2022, 3:45 a.m. |
Created
41
|
[{"model": "core.projectorganisation", "pk": 60681, "fields": {"project": 1073, "organisation": 1855, "role": "PARTICIPANT_ORG"}}]
|
|
| April 11, 2022, 3:45 a.m. |
Created
41
|
[{"model": "core.projectorganisation", "pk": 60680, "fields": {"project": 1073, "organisation": 1855, "role": "LEAD_ORG"}}]
|
|
| April 11, 2022, 3:45 a.m. |
Created
40
|
[{"model": "core.projectperson", "pk": 37384, "fields": {"project": 1073, "person": 2153, "role": "PM_PER"}}]
|
|
| April 11, 2022, 1:47 a.m. |
Updated
35
|
{"title": ["", "LEO Satellite Based AI Demonstrator"], "description": ["", "\nSatellites typically have limited computing power, in part because they are solar powered and because their rigorous testing schedules and inaccessible operating location demands reliable, time proven technology, often several generations behind current state of the art devices we are familiar with.\n\nOur project aims to automatically produce a deep learning, object detection algorithm, which will be compressed and optimised to run on a space-grade FPGA device qualified to work in space on a satellite. The object detection algorithm will use synthetic aperture radar (SAR) and hyper-spectral image data as input sources and it will be trained using existing archives of satellite SAR and image data. The final deep learning system will be tested by Thales Alenia Space, a prime space contractor for ESA, at their satellite facility in Bristol UK.\n\nSatellites periodically transmit large volumes of collected data to earth based receiving stations for processing and distribution. This cyclic process restricts how much data can be collected during an orbit and requires significant bandwidth to transmit and receive data during the downlink window. By enabling the satellite with on-board object detection, it will identify and respond in real-time to observed events and then be selective about which data to source and keep for later downloads.\n\nThese are fundamental problems with current satellite technology. It is relatively easy to attach high resolution scanners and radars to satellites, but much harder to store and transmit the volumes of data that can be gathered during one or more orbits. By finding ways to put smart AI algorithms into the limited, on-board compute devices of satellites we will make more efficient use of their capabilities and in-turn enable satellites and other space vehicles to undertake autonomous activities, when out of communication or too distant from Earth.\n\nThis project is highly innovative because it will automate the design and creation of an object detection algorithm on a minimally configured, space-grade FPGA. If space technology is to reliably exploit AI algorithms this capability will be essential. There are no AI processors currently designed for space use.\n\nAlthough this project is developing an AI solution for a satellite platform, our solution is equally applicable to other space applications on deep space vehicles or on planetary rovers. It would require a different deep learning algorithm, which would need to be re-trained for the specific task, but the same space-grade FPGAs could be used.\n\n"], "extra_text": ["", "\n\n\n\n"], "status": ["", "Closed"]}
|
|
| April 11, 2022, 1:46 a.m. |
Added
35
|
{"external_links": [3821]}
|
|
| April 11, 2022, 1:46 a.m. |
Created
35
|
[{"model": "core.project", "pk": 1073, "fields": {"owner": null, "is_locked": false, "coped_id": "e277a889-fe7a-4368-9be1-a86488a96fa8", "title": "", "description": "", "extra_text": "", "status": "", "start": null, "end": null, "raw_data": 6294, "created": "2022-04-11T01:30:59.460Z", "modified": "2022-04-11T01:30:59.460Z", "external_links": []}}]
|
|