Create Simulink Environment and Train Agent

This example shows how to convert the PI controller in the watertank Simulink® model to a reinforcement learning deep deterministic policy gradient (DDPG) agent. For an example that trains a DDPG agent in MATLAB®, see Train DDPG Agent to Control Double Integrator System (Reinforcement Learning Toolbox).

Water Tank Model

The original model for this example is the water tank model. The goal is to control the level of the water in the tank. For more information about the water tank model, see watertank Simulink Model (Simulink Control Design).

Modify the original model by making the following changes:

1. Delete the PID Controller.

2. Insert the RL Agent block.

3. Connect the observation vector ${\left[\begin{array}{ccc}\int \mathit{e}\mathrm{dt}& \mathit{e}& \mathit{h}\end{array}\right]}^{\mathit{T}}$, where $\mathit{h}$ is the height of the tank, $\mathit{e}=\mathit{r}-\mathit{h}$, and $\mathit{r}$ is the reference height.

4. Set up the reward $\mathrm{reward}=10\left(\left|\mathit{e}\right|<0.1\right)-1\left(\left|\mathit{e}\right|\ge 0.1\right)-100\left(\mathit{h}\le 0\left|\right|\mathit{h}\ge 20\right)$.

5. Configure the termination signal such that the simulation stops if $\mathit{h}\le 0$ or $\mathit{h}\ge 20$.

The resulting model is rlwatertank.slx. For more information on this model and the changes, see Create Simulink Reinforcement Learning Environments (Reinforcement Learning Toolbox).

open_system('rlwatertank')

Create the Environment Interface

Creating an environment model includes defining the following:

• Action and observation signals that the agent uses to interact with the environment. For more information, see rlNumericSpec (Reinforcement Learning Toolbox) and rlFiniteSetSpec (Reinforcement Learning Toolbox).

• Reward signal that the agent uses to measure its success. For more information, see Define Reward Signals (Reinforcement Learning Toolbox).

Define the observation specification obsInfo and action specification actInfo.

obsInfo = rlNumericSpec([3 1],...
    LowerLimit=[-inf -inf 0  ]',...
    UpperLimit=[ inf  inf inf]');
obsInfo.Name = "observations";
obsInfo.Description = "integrated error, error, and measured height";

actInfo = rlNumericSpec([1 1]);
actInfo.Name = "flow";

Build the environment interface object.

env = rlSimulinkEnv("rlwatertank","rlwatertank/RL Agent",...
    obsInfo,actInfo);

Set a custom reset function that randomizes the reference values for the model.

env.ResetFcn = @(in)localResetFcn(in);

Specify the simulation time Tf and the agent sample time Ts in seconds.

Ts = 1.0;
Tf = 200;

Fix the random generator seed for reproducibility.

rng(0)

Create the Critic

Given observations and actions, a DDPG agent approximates the long-term reward using a value function approximator as a critic.

Create a deep neural network to approximate the value function within the critic. To create a network with two inputs, the observation and action, and one output, the value, use three different paths, and specify each path as a row vector of layer objects. You can obtain the dimension of the observation and action spaces from the obsInfo and actInfo specifications.

statePath = [
    featureInputLayer(obsInfo.Dimension(1),Name="netObsIn")
    fullyConnectedLayer(50)
    reluLayer
    fullyConnectedLayer(25,Name="CriticStateFC2")];

actionPath = [
    featureInputLayer(actInfo.Dimension(1),Name="netActIn")
    fullyConnectedLayer(25,Name="CriticActionFC1")];

commonPath = [
    additionLayer(2,Name="add")
    reluLayer
    fullyConnectedLayer(1,Name="CriticOutput")];

criticNetwork = layerGraph();
criticNetwork = addLayers(criticNetwork,statePath);
criticNetwork = addLayers(criticNetwork,actionPath);
criticNetwork = addLayers(criticNetwork,commonPath);

criticNetwork = connectLayers(criticNetwork, ...
    "CriticStateFC2", ...
    "add/in1");
criticNetwork = connectLayers(criticNetwork, ...
    "CriticActionFC1", ...
    "add/in2");

View the critic network configuration.

figure
plot(criticNetwork)

Convert the network to a dlnetwork object and summarize its properties.

criticNetwork = dlnetwork(criticNetwork);
summary(criticNetwork)

   Initialized: true

   Number of learnables: 1.5k

   Inputs:
      1   'netObsIn'   3 features
      2   'netActIn'   1 features

Create the critic approximator object using the specified deep neural network, the environment specification objects, and the names if the network inputs to be associated with the observation and action channels.

critic = rlQValueFunction(criticNetwork,obsInfo,actInfo, ...
    ObservationInputNames="netObsIn", ...
    ActionInputNames="netActIn");

For more information on Q-value function objects, see rlQValueFunction (Reinforcement Learning Toolbox).

Check the critic with a random input observation and action.

getValue(critic, ...
    {rand(obsInfo.Dimension)}, ...
    {rand(actInfo.Dimension)})

ans = single
    -0.1631

For more information on creating critics, see Create Policies and Value Functions (Reinforcement Learning Toolbox).

Create the Actor

Given observations, a DDPG agent decides which action to take using a deterministic policy, which is implemented by an actor.

Create a deep neural network to approximate the policy within the actor. To create a network with one input, the observation, and one output, the action, a row vector of layer objects. You can obtain the dimension of the observation and action spaces from the obsInfo and actInfo specifications.

actorNetwork = [
    featureInputLayer(obsInfo.Dimension(1))
    fullyConnectedLayer(3)
    tanhLayer
    fullyConnectedLayer(actInfo.Dimension(1))
    ];

Convert the network to a dlnetwork object and summarize its properties.

actorNetwork = dlnetwork(actorNetwork);
summary(actorNetwork)

   Initialized: true

   Number of learnables: 16

   Inputs:
      1   'input'   3 features

Create the actor approximator object using the specified deep neural network, the environment specification objects, and the name if the network input to be associated with the observation channel.

actor = rlContinuousDeterministicActor(actorNetwork,obsInfo,actInfo);

For more information, see rlContinuousDeterministicActor (Reinforcement Learning Toolbox).

Check the actor with a random input observation.

getAction(actor,{rand(obsInfo.Dimension)})

ans = 1x1 cell array
    {[-0.3408]}

For more information on creating critics, see Create Policies and Value Functions (Reinforcement Learning Toolbox).

Create the DDPG Agent

Create the DDPG agent using the specified actor and critic approximator objects.

agent = rlDDPGAgent(actor,critic);

For more information, see rlDDPGAgent (Reinforcement Learning Toolbox).

Specify options for the agent, the actor, and the critic using dot notation.

agent.SampleTime = Ts;

agent.AgentOptions.TargetSmoothFactor = 1e-3;
agent.AgentOptions.DiscountFactor = 1.0;
agent.AgentOptions.MiniBatchSize = 64;
agent.AgentOptions.ExperienceBufferLength = 1e6; 

agent.AgentOptions.NoiseOptions.Variance = 0.3;
agent.AgentOptions.NoiseOptions.VarianceDecayRate = 1e-5;

agent.AgentOptions.CriticOptimizerOptions.LearnRate = 1e-03;
agent.AgentOptions.CriticOptimizerOptions.GradientThreshold = 1;
agent.AgentOptions.ActorOptimizerOptions.LearnRate = 1e-04;
agent.AgentOptions.ActorOptimizerOptions.GradientThreshold = 1;

Alternatively, you can specify the agent options using an rlDDPGAgentOptions (Reinforcement Learning Toolbox) object.

Check the agent with a random input observation.

getAction(agent,{rand(obsInfo.Dimension)})

ans = 1x1 cell array
    {[-0.7926]}

Train Agent

To train the agent, first specify the training options. For this example, use the following options:

• Run each training for at most 5000 episodes. Specify that each episode lasts for at most ceil(Tf/Ts) (that is 200) time steps.

• Display the training progress in the Episode Manager dialog box (set the Plots option) and disable the command line display (set the Verbose option to false).

• Stop training when the agent receives an average cumulative reward greater than 800 over 20 consecutive episodes. At this point, the agent can control the level of water in the tank.

For more information, see rlTrainingOptions (Reinforcement Learning Toolbox).

trainOpts = rlTrainingOptions(...
    MaxEpisodes=5000, ...
    MaxStepsPerEpisode=ceil(Tf/Ts), ...
    ScoreAveragingWindowLength=20, ...
    Verbose=false, ...
    Plots="training-progress",...
    StopTrainingCriteria="AverageReward",...
    StopTrainingValue=800);

Train the agent using the train (Reinforcement Learning Toolbox) function. Training is a computationally intensive process that takes several minutes to complete. To save time while running this example, load a pretrained agent by setting doTraining to false. To train the agent yourself, set doTraining to true.

doTraining = false;

if doTraining
    % Train the agent.
    trainingStats = train(agent,env,trainOpts);
else
    % Load the pretrained agent for the example.
    load("WaterTankDDPG.mat","agent")
end

Validate Trained Agent

Validate the learned agent against the model by simulation.

simOpts = rlSimulationOptions(MaxSteps=ceil(Tf/Ts),StopOnError="on");
experiences = sim(env,agent,simOpts);

Local Function

function in = localResetFcn(in)

% randomize reference signal
blk = sprintf('rlwatertank/Desired \nWater Level');
h = 3*randn + 10;
while h <= 0 || h >= 20
    h = 3*randn + 10;
end
in = setBlockParameter(in,blk,'Value',num2str(h));

% randomize initial height
h = 3*randn + 10;
while h <= 0 || h >= 20
    h = 3*randn + 10;
end
blk = 'rlwatertank/Water-Tank System/H';
in = setBlockParameter(in,blk,'InitialCondition',num2str(h));

end