From e9dc24a7a7b7bc80cf3d7c1510239a1191b8d092 Mon Sep 17 00:00:00 2001
From: Andru Liu <90433630+WallabyLester@users.noreply.github.com>
Date: Mon, 7 Oct 2024 23:29:36 -0700
Subject: [PATCH] Updating README.

---
 README.md | 20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)

diff --git a/README.md b/README.md
index 580e36f..5481c1c 100644
--- a/README.md
+++ b/README.md
@@ -1,2 +1,22 @@
 # RBF-aPID-Controller
 RBF Neural Net Adaptive PID Controller
+
+Implementations of a radial basis function (RBF) neural network adaptive PID controller. Uses 
+neural net and error information from PID control to adapt the control signal. Provides one 
+adaptation value, using error, integral, and derivative. 
+
+To adapt the PID gains themselves, network outputs must be made to 3 neurons. Example usage 
+with simulated data can be found in `first_order_sim.py`
+
+The method has been implemented in three ways: 
+
+1. TensorFlow : Using TF to build and train the RBF Model.
+
+![TensorFlow](images/tf_impl.png "TensorFlow")
+![TF_Trained](images/trained.png "TF_Trained")
+
+2. Numpy: Using Numpy to build and train the RBF Model.
+
+![Numpy](images/nump_impl.png "Numpy")
+
+3. C++: Written in C++ (requiring `cmath`), for use on embedded systems.