skills and knowledge list.txt

> 3 week sprint focus proiorities, my A+ problem (each line is the order position for each 3 week block of focus)
To be clear, the first thing in the list has an active project related.
>  SLAM and neural networks applied in vehicles - The comma.ai Calibration Challenge!
> Understand and apply: networking/browsers, from the transistor to google, HTML - Create a personal site
> Use an arduino, servo motors, a driver, and 3d printed parts to do a simple robotics project (learn fusion360, ROS, etcetc)

My greatest A problems, are: important, urgent:
> Python/Numpy/Pygame general software engineering/hacking
> Mastering PyTorch/SciPy, understanding and apply neural networks (model details, loss formulations, algos)
> Understand and apply: Full body locomotion, path planning
> Understand and apply: Object manipulation - grasp, pick-and-place, rotate, opening doors
> Understand and apply: Objection/pose detection (OpenCV)
> Understand and apply: 
> Understand and apply: 
> Understand and apply: Depth estimation
> Understand and apply: fundamental geometric concepts such as perspective transformation, pinhole camera model, and epipolar geometry
> Understand and apply: Model quantization in deep learning 
> Understad and apply: Imitaion learning, and reinforcemnt learning.
> Implement end to end: Sensors, planning+processing, and execution in a real robot.
> Understand Vector and optimisation Calculus, Linear Algebra, Bayesian and otherwise probability, numerical optimisation

The biggest B problems, things I know are useful within 10 years and are: important, not urgent:
> C++
> Experience developing modern planning approaches such as search, sampling, trajectory optimization and model predictive control on real robots.
> For NLP, do an end to end application with a physical Ai
> Understand distributed deep learning
> Understand GPU's/TPU's/AlgoComplexity(time taken)/CUDA for training neural networks
> Understand and apply: Spark, SQL, other data structures
> Understand and apply: Visual inertial odometry
> Understand and apply: Structure from motion
> Understand and apply: 3D reconsctruction
> Understand and apply: Tacticle sensors, camera sensors, moisture/chemical sensors
> Understand and apply: State machines and behaviour trees
> Understand and apply: Beahviour authoring/coordination tools
> Understand and apply: SLAM (online and offline state estimation, fuse info from IMU's, stereo cameras)
> Understanding and apply Kernel, Docket, git, Linux
Tackle problems across multiple domains including but not limited to high level decision-making, multi-agent interactions and trajectory generation
> Undesrtand energy/tech type conversion (file types, energy transfers, the WHOLE CHAIN END TO END BABY)
Knowledge of microcontroller peripherals such as ADC, GPIO, PWM , SPI and I2C
> Understanding image processing and graphics
> 

Things that I have to learn soon but eh and are: not important urgent:
> Java

Things that are: not important, not urgent:
> Continuous Integration/Continuous Deployment (CI/CD) tools like Jenkins or Travis CI

Ideas:
> Autonomus supply chain management, find points of cost and reduce with robotic + managed by NLP autonomous agents/robots.