Steven Omole (@OmoleSteven) profile picture

Roboticist | Researcher | Engineer | Currently building a Robot 😎

Campania, Italy
Joined May 2018
Photo by Steven Omole (@OmoleSteven)Photo by Steven Omole (@OmoleSteven)Photo by Steven Omole (@OmoleSteven)Photo by Steven Omole (@OmoleSteven)Photo by Steven Omole (@OmoleSteven)Photo by Steven Omole (@OmoleSteven)Photo by Steven Omole (@OmoleSteven)Photo by Steven Omole (@OmoleSteven)Photo by Steven Omole (@OmoleSteven)Photo by Steven Omole (@OmoleSteven)
Want to teach a robot to see? Start here. Robot Vision: Zero to SLAM by @Pouya_Mn is a free, bilingual book that takes you from basic images all the way to Visual SLAM and ROS 2. It starts with NumPy and OpenCV, then moves through image processing, feature detection, camera geometry, YOLO, deep learning, visual odometry, SLAM and real-time vision in ROS 2. The nice part is that you only need basic Python to start. Each chapter has runnable examples, so you learn by actually building things instead of just reading theory. Free, open-source, and built to learn by doing.
5
12
75
2,201
Steven Omole retweeted
We release Needle 2: A 14MB agentic LLM for phones, wearables, smart home, robots and microcontroller. The whole model is a single 14MB binary that runs a full session in 28MB of RAM. It is built on our Simple Attention Network findings, compressed to CQ2-bit with Cactus Quants, and baked into its own engine. Needle 2 has 45m parameters trained from the ground up on 140B tool call, device use structured generation tokens. On mobile device use benchmarks, Needle 2 trades wins with frontier small LLMs like LFM2.5 230M, Apple FM Gemma-270m, at 5× to 70× smaller, and 2 bits against their f16. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, between 400–1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges 300–700 on sub-$200 phones such as the Samsung A-Series. Needle also runs on newer microcontrollers like ESP32. A conventional transformer of Needle's width and depth spends 164 MFLOPs per token, and even one squeezed down to Needle's parameter count spends 87, Needle spends 70. Even on a high-end phone, an always-on assistant lives inside a power budget; every MFLOP is milliwatt-hours, and Needle spends 7x to 85x fewer of them per token than the smallest performant LLMs. Read more: cactuscompute.com/needle
100
244
61
2,410
294,836
Meet the Variable Form #Robot That Adapts to Any Terrain via @ZappyZappy7 #Robotics #Engineering #Innovation #Technology #TechForGood
5
49
3
227
14,082
DEEP #Robotics Unveils a Hyper-Realistic #Robot Horse That Moves Like the Real Thing via @XRoboHub #AI #ArtificialIntelligence #EmergingTech #TechForGood
19
78
7,472
Let’s go ✈️
Pets do not have to die anymore. This robot dog tracks your palm, locks two green eyes on you, then walks off on four metal legs. No leash. No remote in the shot. It just stays with you. A real dog gets old. This one does not. When the body wears out you swap the shell. The face, the habits, the way it looks at your hand stay. Same companion. No last day.
9
93
Companion & Social Robots have come to stay. Let’s go ✈️
The next robot you buy might not work in a factory. It might live on your desk. A new wave of small AI robots is turning companionship, personality and physical interaction into real consumer products. • MicroDuck • Palmimo / Mi-Mo • Reachy Mini • EMO • AIBI • Eilik • Loona • Ropet • Moflin • NICOBO • Mirumi • BOCCO emo • EBO X • Vector • Doly • ClicBot Some are pets. Some are companions. Some are open platforms for developers. Which one has the best chance of becoming a real mainstream product?
1
56
The next robot you buy might not work in a factory. It might live on your desk. A new wave of small AI robots is turning companionship, personality and physical interaction into real consumer products. • MicroDuck • Palmimo / Mi-Mo • Reachy Mini • EMO • AIBI • Eilik • Loona • Ropet • Moflin • NICOBO • Mirumi • BOCCO emo • EBO X • Vector • Doly • ClicBot Some are pets. Some are companions. Some are open platforms for developers. Which one has the best chance of becoming a real mainstream product?
This quoted post is unavailable.
14
25
9
167
26,950
Massive week in robotics. I went through everything from Unitree Robotics, Dyna Robotics, Flexion, and even more. Get your coffee and see what happened in robotics space:
9
25
3
316
26,305
Testing the leg limb part of my ongoing humanoid project #Robotics #Humanoid #Servo
2
44
Steven Omole retweeted
one of the quotes i find most inspiring on a hard day: "Whatever your hand finds to do, do it with all your might, for in the realm of the dead, where you are going, there is neither working nor planning nor knowledge nor wisdom" Ecclesiastes 9:10
1,208
2,608
372
19,131
1,842,268
LE TOGO SUPPRIME LE VISA POUR TOUS LES AFRICAINS ! Le Togo franchit une étape historique dans le renforcement de l’intégration africaine. Désormais, tous les ressortissants des États africains détenteurs d’un passeport national valide peuvent entrer sur le territoire togolais sans visa, pour un séjour allant jusqu’à 30 jours. À travers cette réforme majeure, le Président du Conseil réaffirme sa volonté de faire du Togo un espace d’ouverture, de mobilité, d’opportunités et de coopération au cœur du continent africain. Les voyageurs doivent toutefois effectuer leur déclaration de voyage sur la plateforme officielle voyage.gouv.tg au moins 24 heures avant leur arrivée afin d’obtenir leur bordereau de voyage. Le Togo confirme ainsi son leadership en matière d’intégration régionale et de rapprochement des peuples africains. #Togo #Afrique #integration #Libre #panafricanism #voyage #cooperation
306
3,283
447
7,703
584,535
For too long, Africa’s resources have been extracted, the value captured elsewhere, the environmental damage left behind. No more exploitation. No more plundering. The people of Africa must benefit - first & most - from the resources of Africa.
1,082
2,610
453
6,469
258,927
Princeton's Introduction to Robotics! 🎓 @Princeton University released their full Introduction to Robotics course publicly with lecture videos, notes, slides, and assignments. This course provides fundamental theoretical and algorithmic principles behind robotic systems with hands-on experience. Topics covered: → Feedback Control (dynamics, PD control, Linear Quadratic Regulator) → Motion Planning (discrete planning with BFS/DFS, optimal planning with Dijkstra/A*) → State Estimation, Localization, and Mapping (Bayes filtering, Kalman filtering, particle filtering, SLAM) → Vision and Learning (optical flow, deep learning, convolutional networks, reinforcement learning), and broader topics including robotics and law, ethics, and economics. Assignments include theory, programming, and hardware implementation components. The final project has students program drones for vision-based navigation with attached cameras transmitting real-time images. All lecture videos, notes, slides, and assignments are freely available. Prerequisites include multivariable calculus, linear algebra, basic probability, basic differential equations, and some programming experience in Python. ‼️ GO FOR IT: irom-lab.princeton.edu/intro… ~~ ♻️ Join the weekly robotics newsletter, and never miss any news → ziegler.substack.com
8
210
3
1,244
52,992
Steven Omole retweeted
A full MIT course on robot mechanics and control. If you're building or working with robotic systems, this one deserves a permanent bookmark.📌 Russ Tedrake's Underactuated Robotics at MIT covers the math and intuition behind how robots actually move — not just the surface level. No paywalls. No prerequisites gatekeeping. What it focuses on: - Nonlinear dynamics and stability for robotic systems - Trajectory optimization and motion planning - Reinforcement learning applied to locomotion and manipulation - Lyapunov methods and limit cycles for real control design - Worked examples with code across the full course Full textbook, lecture videos, and problem sets — all free. 📍 underactuated.mit.edu
8
147
4
841
33,868
Did someone say a free end-to-end robotics workshop with the SO-101 robot? 👀 Join us on Wednesday, April 22 at 11 AM PT for Robotics Office Hours, where NVIDIA experts will demo an end-to-end robot learning workflow using NVIDIA Isaac Sim, Isaac Lab, Cosmos, and Omniverse for sim-to-real manipulation. Learn how to get started. 🦾 📆 Add to calendar: nvda.ws/4sLI50q
2
20
2
155
14,957
Steven Omole retweeted
Second terms are for leaders who deliver. Mediocrity should not be rewarded with more responsibility.
136
11,565
161
24,338
252,344
Steven Omole retweeted
Advanced Robotics: UC Berkeley This is course is from Peter Abbeel and covers a review on reinforcement learning and continues to applications in robotics. If you work on robotics… this one is worth bookmarking‼️ MDPs: Exact Methods Discretization of Continuous State Space MDPs Function Approximation / Feature-based Representations LQR, iterative LQR / Differential Dynamic Programming ... 📍Link to Course: youtube.com/playlist?list=PL… 📍Link to Materials: people.eecs.berkeley.edu/~pa… —- Weekly robotics and AI insights. Subscribe free: scalingdeep.tech
3
80
1
563
26,180
Done
This Post is from a suspended account. Learn more
1
24
You have to take as many calculated risks as possible or else your life will continue to look the same for years.
87
2,489
74
10,039
210,278