Reinforcement learning

22 boards

KiCad boards from 7 GitHub repositories tagged “reinforcement-learning”, the most starred first; open any of them in PCBKit.

2 layers20 × 20 mm
Voltage datalogger adc0
LukeVassallo/RL_PCB
RL_PCB is a novel learning-based method for optimising the placement of circuit components on a Printed Circuit Board (PCB).
9 parts76MIT
2 layers80.5 × 99.9 mm
Automative Hexapod
NguyenTrongPhuc552003/reinforcement-autonomous-hexapod
Intelligent hexapod control system with AI-based adaptive gait, kernel drivers, and user-space coordination.
160 parts3GPL-2.0
4 layers57 × 47.5 mm
Rp2040 debugger
elcruzo/learned-pcb-placement
GNN-guided simulated annealing for PCB component placement. We train a Graph Attention Network on SA rollout data to learn which components to move and where, then use these predictions to bias the SA search.
69 parts2MIT
2 layers73 × 54 mm
FishCoreBoard v3 3
getriscvyy/divesync-heavy
A syringe-actuated buoyancy control device developed at the Bio-Inspired Robotics and Controls Lab at University of Houston.
46 parts1
4 layers120 × 90 mm
Robot platform
telleroutlook/3we-robot-platform
3we Universal Modular Mobile Robot Platform — Open-source firmware, ROS2, hardware, and SDK
8 parts1Apache-2.0
2 layers70 × 50 mm
Monitor
anton-pt/hwdevsbx
RealityCheck: executable hardware design sandboxes, peripheral emulators and an unsaturated benchmark for hardware agents
43 parts
2 layers118 × 125 mm
Kenny carrier revA placement
EruzaDev/Kenny---Automotive-self-driving-trash-can
Open kennycarrierrevA.kicadpro in KiCad 7 or later.
63 parts
2 layers20 × 20 mm
Bistable oscillator with 555 timer and ldo 2lyr setup 00
LukeVassallo/RL_PCB
RL_PCB is a novel learning-based method for optimising the placement of circuit components on a Printed Circuit Board (PCB).
6 parts76MIT
2 layers20 × 20 mm
PModBoard
LukeVassallo/RL_PCB
RL_PCB is a novel learning-based method for optimising the placement of circuit components on a Printed Circuit Board (PCB).
5 parts76MIT
2 layers20 × 20 mm
Tc logger max232
LukeVassallo/RL_PCB
RL_PCB is a novel learning-based method for optimising the placement of circuit components on a Printed Circuit Board (PCB).
5 parts76MIT
2 layers20 × 20 mm
Tc logger max31856
LukeVassallo/RL_PCB
RL_PCB is a novel learning-based method for optimising the placement of circuit components on a Printed Circuit Board (PCB).
8 parts76MIT
2 layers20 × 20 mm
Tc logger mcu
LukeVassallo/RL_PCB
RL_PCB is a novel learning-based method for optimising the placement of circuit components on a Printed Circuit Board (PCB).
4 parts76MIT
2 layers20 × 20 mm
Tc logger silabs
LukeVassallo/RL_PCB
RL_PCB is a novel learning-based method for optimising the placement of circuit components on a Printed Circuit Board (PCB).
8 parts76MIT
2 layers20 × 20 mm
Voltage datalogger adc2
LukeVassallo/RL_PCB
RL_PCB is a novel learning-based method for optimising the placement of circuit components on a Printed Circuit Board (PCB).
3 parts76MIT
2 layers20 × 20 mm
Voltage datalogger afe
LukeVassallo/RL_PCB
RL_PCB is a novel learning-based method for optimising the placement of circuit components on a Printed Circuit Board (PCB).
9 parts76MIT
2 layers110.1 × 111.8 mm
Automative Hexapod
NguyenTrongPhuc552003/reinforcement-autonomous-hexapod
Intelligent hexapod control system with AI-based adaptive gait, kernel drivers, and user-space coordination.
70 parts3GPL-2.0
2 layers120 × 27.3 mm
Keyboard rev0
elcruzo/learned-pcb-placement
GNN-guided simulated annealing for PCB component placement. We train a Graph Attention Network on SA rollout data to learn which components to move and where, then use these predictions to bias the SA search.
37 parts2MIT
2 layers87.6 × 30.5 mm
Mmdvm hotspot
elcruzo/learned-pcb-placement
GNN-guided simulated annealing for PCB component placement. We train a Graph Attention Network on SA rollout data to learn which components to move and where, then use these predictions to bias the SA search.
21 parts2MIT
2 layers25.9 × 17.3 mm
Snapvcc
elcruzo/learned-pcb-placement
GNN-guided simulated annealing for PCB component placement. We train a Graph Attention Network on SA rollout data to learn which components to move and where, then use these predictions to bias the SA search.
26 parts2MIT
2 layers50 × 40 mm
Monitor
anton-pt/hwdevsbx
RealityCheck: executable hardware design sandboxes, peripheral emulators and an unsaturated benchmark for hardware agents
29 parts
2 layers49 × 39 mm
Monitor
anton-pt/hwdevsbx
RealityCheck: executable hardware design sandboxes, peripheral emulators and an unsaturated benchmark for hardware agents
39 parts
2 layers44.9 × 34.9 mm
Workspace
anton-pt/hwdevsbx
RealityCheck: executable hardware design sandboxes, peripheral emulators and an unsaturated benchmark for hardware agents
14 parts