diff --git a/roboteam_ai/src/RL/getRefereeState.py b/roboteam_ai/src/RL/getRefereeState.py
new file mode 100644
index 000000000..4b9fddf05
--- /dev/null
+++ b/roboteam_ai/src/RL/getRefereeState.py
@@ -0,0 +1,72 @@
+import os
+import sys
+import socket
+import struct
+import binascii
+from google.protobuf.json_format import MessageToJson
+
+'''
+getRefereeState.py is a script to get state of the referee. 
+This includes the current command, designed position for ball placement, and the score for both teams.
+'''
+
+# Make sure to go back to the main roboteam directory
+current_dir = os.path.dirname(os.path.abspath(__file__))
+roboteam_path = os.path.abspath(os.path.join(current_dir, "..", "..", ".."))
+
+# Add to sys.path
+sys.path.append(roboteam_path)
+
+# Now import the generated protobuf classes
+from roboteam_networking.proto.ssl_gc_referee_message_pb2 import Referee
+from roboteam_networking.proto.ssl_gc_game_event_pb2 import GameEvent
+
+MULTICAST_GROUP = '224.5.23.1'
+MULTICAST_PORT = 10003
+
+def get_referee_state():
+    # Create the socket
+    sock = socket.socket(socket.AF_INET, socket.SOCK_DGRAM, socket.IPPROTO_UDP)
+    sock.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1)
+    
+    # Bind to the server address
+    sock.bind(('', MULTICAST_PORT))
+    
+    # Tell the operating system to add the socket to the multicast group
+    group = socket.inet_aton(MULTICAST_GROUP)
+    mreq = struct.pack('4sL', group, socket.INADDR_ANY)
+    sock.setsockopt(socket.IPPROTO_IP, socket.IP_ADD_MEMBERSHIP, mreq)
+    
+    print(f"Listening for Referee messages on {MULTICAST_GROUP}:{MULTICAST_PORT}")
+    
+    command = None
+    pos_x = None
+    pos_y = None
+    yellow_score = None
+    blue_score = None
+    
+    try:
+        data, _ = sock.recvfrom(4096)  # Increased buffer size to 4096 bytes
+        referee = Referee()
+        referee.ParseFromString(data)
+        command = Referee.Command.Name(referee.command)
+        
+        if referee.HasField('designated_position'):
+            pos_x = referee.designated_position.x
+            pos_y = referee.designated_position.y
+        
+        yellow_score = referee.yellow.score
+        blue_score = referee.blue.score
+    except Exception as e:
+        print(f"Error parsing message: {e}")
+    finally:
+        sock.close()
+    
+    return command, pos_x, pos_y, yellow_score, blue_score
+
+if __name__ == "__main__":
+    command, pos_x, pos_y, yellow_score, blue_score = get_referee_state()
+    print(f"Command: {command}")
+    print(f"Designated Position: ({pos_x}, {pos_y})")
+    print(f"Yellow Team Score: {yellow_score}")
+    print(f"Blue Team Score: {blue_score}")
\ No newline at end of file
diff --git a/roboteam_ai/src/RL/receiveActionCommand.cpp b/roboteam_ai/src/RL/receiveActionCommand.cpp
new file mode 100644
index 000000000..1a11dc588
--- /dev/null
+++ b/roboteam_ai/src/RL/receiveActionCommand.cpp
@@ -0,0 +1,40 @@
+#include <zmq.hpp>
+#include <string>
+#include <iostream>
+#include <vector>
+#include "ActionCommand.pb.h"
+
+int main() {
+    zmq::context_t context(1);
+    zmq::socket_t socket(context, ZMQ_SUB);
+
+    std::cout << "Connecting to ActionCommand sender..." << std::endl;
+    socket.connect("tcp://localhost:5555");
+    socket.setsockopt(ZMQ_SUBSCRIBE, "", 0);
+
+    std::cout << "ActionCommand receiver started. Ctrl+C to exit." << std::endl;
+
+    while (true) {
+        zmq::message_t message;
+        socket.recv(&message);
+
+        ActionCommand action_command;
+        if (!action_command.ParseFromArray(message.data(), message.size())) {
+            std::cerr << "Failed to parse ActionCommand." << std::endl;
+            continue;
+        }
+
+        if (action_command.numrobots_size() != 3) {
+            std::cerr << "Received incorrect number of values. Expected 3, got " 
+                      << action_command.numrobots_size() << std::endl;
+            continue;
+        }
+
+        std::cout << "Received: [" 
+                  << action_command.numrobots(0) << ", "
+                  << action_command.numrobots(1) << ", "
+                  << action_command.numrobots(2) << "]" << std::endl;
+    }
+
+    return 0;
+}
\ No newline at end of file
diff --git a/roboteam_ai/src/RL/resetRefereeState.py b/roboteam_ai/src/RL/resetRefereeState.py
new file mode 100644
index 000000000..f57f832bc
--- /dev/null
+++ b/roboteam_ai/src/RL/resetRefereeState.py
@@ -0,0 +1,51 @@
+"""
+resetRefereeState is a script to reset the referee state of the game (it basically resets the match (clock))
+"""
+
+import asyncio
+import websockets
+from google.protobuf.json_format import MessageToJson, Parse
+import os
+import sys
+
+# Make sure to go back to the main roboteam directory
+current_dir = os.path.dirname(os.path.abspath(__file__))
+roboteam_path = os.path.abspath(os.path.join(current_dir, "..", "..", ".."))
+
+# Add to sys.path
+sys.path.append(roboteam_path)
+
+# Now import the generated protobuf classes
+from roboteam_networking.proto.ssl_gc_api_pb2 import Input
+from roboteam_networking.proto.ssl_gc_change_pb2 import Change
+from roboteam_networking.proto.ssl_gc_common_pb2 import Team
+from roboteam_networking.proto.ssl_gc_state_pb2 import Command
+
+
+async def reset_and_stop_match(uri='ws://localhost:8081/api/control'):
+    async with websockets.connect(uri) as websocket:
+        
+        # Step 1: Reset the match
+        reset_message = Input(reset_match=True)
+        await websocket.send(MessageToJson(reset_message))
+        response = await websocket.recv()
+
+        # Step 2: Send STOP command
+        stop_message = Input(
+            change=Change(
+                new_command_change=Change.NewCommand(
+                    command=Command(
+                        type=Command.Type.STOP,
+                        for_team=Team.UNKNOWN
+                    )
+                )
+            )
+        )
+        await websocket.send(MessageToJson(stop_message))
+        print(f"Sent STOP command: {MessageToJson(stop_message)}")
+        response = await websocket.recv()
+
+        print("Reset and STOP commands sent to SSL Game Controller")
+
+if __name__ == "__main__":
+    asyncio.run(reset_and_stop_match())
\ No newline at end of file
diff --git a/roboteam_ai/src/RL/train.py b/roboteam_ai/src/RL/train.py
new file mode 100644
index 000000000..467a421d4
--- /dev/null
+++ b/roboteam_ai/src/RL/train.py
@@ -0,0 +1,24 @@
+import gymnasium as gym
+from stable_baselines3 import PPO
+
+# Import your custom environment
+from env import RoboTeamEnv
+
+# Create the environment
+env = RoboTeamEnv()
+
+# Create and train the PPO model
+model = PPO("MultiInputPolicy", env, verbose=1)
+model.learn(total_timesteps=1000)
+
+# Test the trained model
+obs, _ = env.reset()
+for i in range(1000):
+    action, _states = model.predict(obs, deterministic=True)
+    obs, reward, terminated, truncated, info = env.step(action)
+    env.render()
+    if terminated or truncated:
+        obs, _ = env.reset()
+
+env.close()
+
diff --git a/roboteam_mpi/README.md b/roboteam_mpi/README.md
new file mode 100644
index 000000000..c8f3fc57b
--- /dev/null
+++ b/roboteam_mpi/README.md
@@ -0,0 +1,13 @@
+### Explanation
+roboteam_mpi is meant to house all the communication for MPI (message processing interface) to work on HPC clusters.
+
+
+This will be a framework where other teams can attach their AI to.
+
+
+## Core components
+
+# MPIManager
+Handles initialization, finalization and standard operations
+
+
diff --git a/roboteam_mpi/mpi_combined.py b/roboteam_mpi/mpi_combined.py
new file mode 100644
index 000000000..7d782f99e
--- /dev/null
+++ b/roboteam_mpi/mpi_combined.py
@@ -0,0 +1,28 @@
+# mpi_combined.py
+from mpi4py import MPI
+import sys
+
+def main():
+    comm = MPI.COMM_WORLD
+    rank = comm.Get_rank()
+    size = comm.Get_size()
+
+    print(f"Process {rank}: I am rank {rank} out of {size} processes")
+    sys.stdout.flush()
+
+    if rank == 0:
+        number = 42
+        print(f"Process {rank}: Sending number {number} to rank 1")
+        sys.stdout.flush()
+        comm.send(number, dest=1, tag=11)
+        print(f"Process {rank}: Number {number} sent to rank 1")
+        sys.stdout.flush()
+    elif rank == 1:
+        print(f"Process {rank}: Waiting to receive number from rank 0")
+        sys.stdout.flush()
+        number = comm.recv(source=0, tag=11)
+        print(f"Process {rank}: Received number {number} from rank 0")
+        sys.stdout.flush()
+
+if __name__ == "__main__":
+    main()
\ No newline at end of file
diff --git a/roboteam_mpi/mpi_message b/roboteam_mpi/mpi_message
new file mode 100755
index 000000000..6a8e71955
Binary files /dev/null and b/roboteam_mpi/mpi_message differ
diff --git a/roboteam_mpi/mpi_message.cpp b/roboteam_mpi/mpi_message.cpp
new file mode 100644
index 000000000..503ab94e0
--- /dev/null
+++ b/roboteam_mpi/mpi_message.cpp
@@ -0,0 +1,38 @@
+#include <mpi.h>
+#include <iostream>
+
+int main(int argc, char *argv[]){
+
+MPI_Init(&argc, &argv);
+
+int world_size; //World size is total amount of processes
+MPI_Comm_size(MPI_COMM_WORLD, &world_size);
+
+int world_rank;
+MPI_Comm_rank(MPI_COMM_WORLD, &world_rank);
+
+if (world_rank == 0)
+{
+// Sending a message
+const int message = 42;
+MPI_Send(&message, //
+1, 
+MPI_INT, 
+1, 
+0, 
+MPI_COMM_WORLD);
+
+std::cout << "Process 0 sends number " << message << " to process 1\n";
+}
+else if (world_rank == 1)
+{
+// Receiving a message
+int received_message;
+MPI_Recv(&received_message, 1, MPI_INT, 0, 0, MPI_COMM_WORLD, MPI_STATUS_IGNORE);
+std::cout << "Process 1 received number " << received_message << " from process 0\n";
+}
+
+MPI_Finalize();
+return 0;
+
+}
\ No newline at end of file
diff --git a/roboteam_mpi/mpi_receiver.py b/roboteam_mpi/mpi_receiver.py
new file mode 100644
index 000000000..59cb0c90a
--- /dev/null
+++ b/roboteam_mpi/mpi_receiver.py
@@ -0,0 +1,14 @@
+from mpi4py import MPI
+
+def main():
+    comm = MPI.COMM_WORLD
+    rank = comm.Get_rank()
+
+    if rank == 1:
+        while True:
+            message = comm.recv(source=0, tag=11)
+            print(f"Receiver: {message}")
+
+if __name__ == "__main__":
+    main()
+    
\ No newline at end of file
diff --git a/roboteam_mpi/mpi_sender.py b/roboteam_mpi/mpi_sender.py
new file mode 100644
index 000000000..bd14fbd4a
--- /dev/null
+++ b/roboteam_mpi/mpi_sender.py
@@ -0,0 +1,12 @@
+from mpi4py import MPI
+
+def main():
+    comm = MPI.COMM_WORLD
+    rank = comm.Get_rank()
+
+    if rank == 0:
+        message = "Test"
+        comm.send(message, dest=1, tag=11)
+        print(f"Sender: {message}")
+if __name__ == "__main__":
+    main()
\ No newline at end of file
diff --git a/roboteam_mpi/mpi_test.py b/roboteam_mpi/mpi_test.py
new file mode 100644
index 000000000..9eea25c76
--- /dev/null
+++ b/roboteam_mpi/mpi_test.py
@@ -0,0 +1,7 @@
+from mpi4py import MPI
+
+comm = MPI.COMM_WORLD
+rank = comm.Get_rank()
+size = comm.Get_size()
+
+print(f"Hello from process {rank} out of {size} processes")
\ No newline at end of file
diff --git a/roboteam_mpi/src/MPIManager.cpp b/roboteam_mpi/src/MPIManager.cpp
new file mode 100644
index 000000000..60e41b04d
--- /dev/null
+++ b/roboteam_mpi/src/MPIManager.cpp
@@ -0,0 +1,37 @@
+#include "MPIManager.h"
+#include <stdexcept>
+
+bool MPIManager::initialized = false;
+int MPIManager::world_rank;
+int MPIManager::world_size;
+
+void MPIManager::init(int& argc, char**& argv) {
+    if (!initialized) {
+        MPI_Init(&argc, &argv);
+        MPI_Comm_rank(MPI_COMM_WORLD, &world_rank); // Unique ID for every process
+        MPI_Comm_size(MPI_COMM_WORLD, &world_size); // Total number of processes
+        initialized = true;
+    }
+}
+
+void MPIManager::finalize() {
+    if (initialized) {
+        MPI_Finalize();
+        initialized = false;
+    }
+}
+
+int MPIManager::getRank() {
+    if (!initialized) throw std::runtime_error("MPI not initialized");
+    return rank;
+}
+
+int MPIManager::getSize() {
+    if (!initialized) throw std::runtime_error("MPI not initialized");
+    return size;
+}
+
+void MPIManager::send(const void* data, int count, MPI_Datatype datatype, int dest, int tag) {
+    if (!initialized) throw std::runtime_error("MPI not initialized");
+    MPI_Send(data, count, datatype, dest, tag, MPI_COMM_WORLD);
+}
diff --git a/roboteam_mpi/src/MPIManager.h b/roboteam_mpi/src/MPIManager.h
new file mode 100644
index 000000000..75a365a89
--- /dev/null
+++ b/roboteam_mpi/src/MPIManager.h
@@ -0,0 +1,25 @@
+#ifndef MPI_MANAGER_H
+#define MPI_MANAGER_H
+
+#include <mpi.h>
+#include <vector>
+#include <string>
+
+class MPIManager {
+public:
+    static void init(int& argc, char**& argv);
+    static void finalize();
+    static int getRank();
+    static int getSize();
+    
+    static void send(const void* data, int count, MPI_Datatype datatype, int dest, int tag);
+    static void recv(void* data, int count, MPI_Datatype datatype, int source, int tag);
+    static void bcast(void* data, int count, MPI_Datatype datatype, int root);
+
+private:
+    static bool initialized;
+    static int rank;
+    static int size;
+
+    MPIManager() = delete;  // Prevent instantiation
+};
\ No newline at end of file
diff --git a/ssl-game-controller b/ssl-game-controller
new file mode 160000
index 000000000..feb9b7636
--- /dev/null
+++ b/ssl-game-controller
@@ -0,0 +1 @@
+Subproject commit feb9b76361f30a9e9b476bc1f52516df0f33de17