Warning: An error occurred while drawing the scene: Error in json_scenetree: Could not find node in replaceChild

My matlab version is 2020(a).

It always calls the "Warning: An error occurred while drawing the scene: Error in json_scenetree: Could not find node in replaceChild" .

I had try the previous solution "opengl software" , but it didn't work.

This is the code I complie.

Is the code have problem or matlab problem.

How can I deal with this problem?

clear all
close all
clc

ATT = 1;
DATE = datestr(now, 30);

mdl = 'quarter_car';
open_system(mdl)
agentblk = [mdl '/RL Agent'];

numObs = 5;
numAct = 2;


observationInfo = rlNumericSpec([numObs 1],'LowerLimit',-inf*ones(numObs,1),'UpperLimit',inf*ones(numObs,1));
observationInfo.Name = 'observation';

actionInfo = rlFiniteSetSpec({[0;0],[0;1]});
actionInfo.Name = 'actor';

env = rlSimulinkEnv(mdl,agentblk,observationInfo,actionInfo);

Ts = 0.01;
Tf = 10;

rng(0)

statePath = [
    imageInputLayer([numObs 1 1],'Normalization','none','Name','observation')
    fullyConnectedLayer(50,'Name','CriticObsFC1')
    reluLayer('Name','CriticRelu1')
    fullyConnectedLayer(30,'Name','CriticObsFC2')];
actionPath = [
    imageInputLayer([numAct 1 1],'Normalization','none','Name','action')
    fullyConnectedLayer(30,'Name','CriticActFC1')];
commonPath = [
    additionLayer(2,'Name','add')
    reluLayer('Name','CriticCommonRelu')
    fullyConnectedLayer(1,'Name','output')];

criticNetwork = layerGraph(statePath);
criticNetwork = addLayers(criticNetwork,actionPath);
criticNetwork = addLayers(criticNetwork,commonPath);    
criticNetwork = connectLayers(criticNetwork,'CriticObsFC2','add/in1');
criticNetwork = connectLayers(criticNetwork,'CriticActFC1','add/in2');

criticOpts = rlRepresentationOptions('LearnRate',0.001,'GradientThreshold',1);
obsInfo = getObservationInfo(env);
actInfo = getActionInfo(env);
critic = rlQValueRepresentation(criticNetwork,obsInfo,actInfo,'Observation',{'observation'},'Action',{'action'},criticOpts);

agentOptions = rlDQNAgentOptions(...
    'SampleTime',Ts,...
    'TargetSmoothFactor',1e-3,...
    'ExperienceBufferLength',1e6,... 
    'UseDoubleDQN',false,...
    'DiscountFactor',0.5,...
    'MiniBatchSize',64);

agentOpts.NoiseOptions.Variance = 0.5;
agentOpts.NoiseOptions.VarianceDecayRate = 1e-3; 

agent = rlDQNAgent(critic,agentOptions);

maxepisodes = 2000;
maxsteps = ceil(Tf/Ts);

trainingOptions = rlTrainingOptions(...
    'MaxEpisodes',maxepisodes,...
    'MaxStepsPerEpisode',maxsteps,...
    'ScoreAveragingWindowLength',5,...
    'Verbose',false,...
    'Plots','training-progress',...
    'StopTrainingCriteria','AverageReward',...
    'StopTrainingValue', 25000);

trainingStats = train(agent,env,trainingOptions);

simOptions = rlSimulationOptions('MaxSteps', maxsteps);
experience = sim(env,agent,simOptions);

Answer 1

[Emmanouil Tzorakoleftherakis]

My suspicion is that the error does not have much to do with the code you are showing but with how you create your environment. It seems you are using Simulink here - are you plotting anything in your Simulink model? Maybe trying to visualize what is happening with your system? That's what I would check first.