[1] National Geographic artikkeli apinoiden orjuutuksesta :
Monkeys still forced to pick coconuts in Thailand despite controversy
"Months after reports that Thailand coconut farmers rely on monkey labor, a new investigation shows little action has been taken." -- Rachel Fobar 19.02.2021, https://www.nationalgeographic.com/animals/article/monkey-labor-continues-in-thailands-coconut-market
[2] konenäkö kykenee erottelemaan omenan kypsyyttä:
"pyTorch [...] We made use of mobile phones to create four datasets [...] with a total of 4,000 images and 20,000 labels. [...] From the experimental results, we observe that both YOLOv8 and CenterNet can achieve an accuracy rate of more than 90% […] the lightweight model YOLOv8n requiring only 2.9ms to complete accurate detection."
-- Fruit ripeness identification using YOLOv8 model. Xiao, B., Nguyen, M. & Yan, W.Q.. https://doi.org/10.1007/s11042-023-16570-9
[3.0] "OpenDR provides the means to link the robotics applications to software libraries (deep learning frameworks, e.g., PyTorch and Tensorflow) to the operating environment (ROS)." -- A modular, open and non-proprietary toolkit for core robotic functionalities by harnessing deep learning, https://github.com/opendr-eu/opendr
[3.1] Yllä mainittu OpenDR on tuottanut tutkimuksessaan konenäön sovelluksen, joka voi navigoida kasveja tunnistamalla ja jota ihminen voi avustaa.
"When the farmer is in the field and can see that the robot is too close to the crop, he is able to nudge the robot to the left or right. Using the same system, the row guidance framework can send the nudge information to the the steering controller, which then adjusts the robot’s trajectory as necessary." --Deliverable D4.3: Third report on deep environment active perception and cognition, Contributing Partners: AU, AUTH, ALU-FR, TUD, TAU,
AGI 31.12.2022,
https://opendr.eu/wp-content/uploads/2023/01/OpenDR_D4.3.pdf
[3.2] RTK-GNSS eli "Real-Time Kinematic positioning using Global Navigation Satellite System", eli GPS/GLONASS/BNS/Galileo -paikannusjärjestelmät varmentaa sijainnin, vaikka kamerallakin voi koittaa navigoida.
"Using the output of the Crop and Weed tool, the crop rows are [detected]. The agriculture field robot is then able to navigate, using the cross track error as a correction to its current position. The actual position is mapped using the robot’s RTK GNSS" -- Deliverable D4.4: Final report on deep environment active perception and cognition, Contributing Partners: AU, AUTH, ALU-FR, TUD, TAU, AGI 29.9.2023, https://opendr.eu/wp-content/uploads/2024/02/D4.4-OpenDR.pdf
[4] Tekoälyn treenaaminen ja simulaatio virtuaalisessa ympäristössä:
"This work proposed an assembly pose estimation method
for robot manipulation [...] The work also presented a synthetic
data generation pipeline as an improvement to existing object
pose estimation datasets." -- 6D Assembly Pose Estimation by Point Cloud Registration for Robot Manipulation
K. Samarawickrama, G. Sharma, A. Angleraud, R. Pieters. arXiv:2312.02593 [cs.RO]
[5] RGB-D, eli "Red/Green/Blue-Depth". Luento-materiaaleista poimittu idea:
"Robot object grasping [...] Perception from camera (RGB-D) to learn grasp poses" -- ROBO.100 / L19 Robotics and AI Research Slides, Roel Pieters, 4-8 Nov 2024, https://moodle.tuni.fi/pluginfile.php/4858298/mod_resource/content/1/ROB_100_Robotics_Research_2024.pdf sivu 17.
[6] "The coolest part? [Our dog shaped robot named] Spot made his way through the apple orchard and even grabbed a few apples. Who knew robots could be farmhands? And guess what? Spot even had an encounter with a real dog. They hit it off right away and became fast friends." -- From Robot Dogs to Cow Buddies: Highlights from AhlmanEdu’s AI Forum, Team Robostudio at AhlmanEdu, Vero, 9.9.2024.
https://blogs.tuni.fi/cs/other-events/from-robot-dogs-to-cow-buddies-highlights-from-ahlmanedus-ai-forum/
[7] Tai ainakin kielten opetuksesta on hyviä kokemuksia tämän lähteen kokeissa. Oma pohdinto: robottimme voisi opettaa vaikka oman laitteistonsa huoltoa, jotta käyttäjät oppisivat itse korjaamaan ja ylläpitämään robottejaan?
Sosiaaliset robotit oppimisen tukena lapsilla ja aikuisilla, Ahtinen, Aino 9.2.2022. https://jyx.jyu.fi/handle/123456789/79736
[8] Lähde: kuulimme IP44 -standardista kerho-huoneella arvostetulta vanhemmalta oppilaalta.🍾🧑🎓
[9] "Hydraulic actuators have [...] several disadvantages: [including] Even the best hydraulic system will leak eventually. [...] A higher noise level" -- Luento-diat L3-L4, L3-L4_2023090406RobotManipulatorsLastra.pdf , sivu 12
[10] Walt Disney Imagineering Research & Development,
Interactive Design of Stylized Walking Gaits for Robotic Characters
July 28, 2024 ACM SIGGRAPH 2024, https://la.disneyresearch.com/publication/interactive-design-of-stylized-walking-gaits-for-robotic-characters/
Katso video:
https://www.youtube.com/watch?v=FDgW3983Cvc
[11] Manipulaattoreille:
Kurssi: ROBO.100-2024 Robotiikan johdantokurssi, Osio: 2. Manipulaattorit | TUNI Moodle
[12] "LUMI (Large Unified Modern Infrastructure) on ainutlaatuinen yhteiseurooppalainen supertietokone, joka sijaitsee CSC:n Kajaanin datakeskuksessa."
-- LUMI: Yksi maailman tehokkaimpia supertietokoneita, CSC – Tieteen tietotekniikan keskus.
https://csc.fi/osaamisemme/suurteholaskenta/lumi-supertietokone/
[13] Raktorille:
Kurssi: ROBO.100-2024 Robotiikan johdantokurssi, Osio: 3. Liikkuvat työkonerobotit | TUNI Moodle
[14] Sopivia EU-rahastoja voisi olla esimerkiksi Horizon Europe (tutkimus ja innovaatio -vaiheissa), Digital Europe Programme (Premium-mallin tuotannon laajennus), tai LIFE Programme (ilmasto ja ympäristö). Erilaisia rahastoja on monenlaisia, kuten EU:n virallisillakin sivuilla todetaan:
"EU funding comes in many different forms:[...]" -- European Union funding, grants and subsidies, https://european-union.europa.eu/live-work-study/funding-grants-subsidies_en