Problem with single agent Simulink using RL toolbox

3 次查看（过去 30 天）
ali farid
ali farid 2024-8-6
编辑： Shantanu Dixit 2024-8-7
I am using RL toolbox to train a single agent with the following specifications:
for type=1
% obsMat = [1 1];
obsMat = [4 3; 5 3; 6 3; 7 3; 8 3; 9 3; 5 11; 6 11; 7 11; 8 11; 6 12; 7 12; 3 12; ];
% obsMat = [ ];
sA0 = [2 5];
switch type
case 1
s0 = [sA0];
end
Ts = 0.1;
Tf = 100;
maxsteps = ceil(Tf/Ts);
switch type
case 1
mdl = "rlAreaCoverage1";
end
open_system(mdl)
% Define observation specifications.
obsSize = [12 12 4];
oinfo = rlNumericSpec(obsSize);
oinfo.Name = "observations";
% Define action specifications.
numAct=131
switch numAct
case 131
actionSpace = {1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 129 130 131 132 133 134 136 137 140 141 142 143 144};
end
ainfo = rlFiniteSetSpec(actionSpace);
ainfo.Name = "actions";
switch type
case 1
blks = mdl + ["/Agent A (Red)"];
end
switch type
case 1
env = rlSimulinkEnv(mdl,blks,{oinfo},{ainfo});
end
env.ResetFcn = @(in) resetMap(in, obsMat);
rng(0)
but I see this error:
Error using rlSimulinkEnv>localValidateIOSpecs (line 178)
Invalid size or type for observation specification.
Error in rlSimulinkEnv (line 90)
localValidateIOSpecs(numAgents,observationInfo,actionInfo);
Do you have any idea what is wrong in my setup?

请先登录，再回答此问题。

回答（1 个）

Shantanu Dixit
Shantanu Dixit 2024-8-7
编辑：Shantanu Dixit 2024-8-7
Hi Ali, it seems that you are facing an issue while training a single agent using RL toolbox in Simulink. Following could be the possible steps to troubleshoot the issue.
  • Correct the initialization of simulink environment as
env = rlSimulinkEnv(mdl,blks,oinfo,ainfo);
  • For a single agent block ensure that the block is defined as a string or character vector containing the block path.
blks = mdl + "rlAreaCoverage1"
  • Ensure that the action space defined matches the action space in the model.
  • Verify that the defined observation size matches the dimensions expected by the model.
  • Verify the output dimensions of the model and ensure that they match with the above specified dimensions.
Additionaly refer to the below MathWorks documentation to learn how to create environment object from a Simulink model

类别

在 Help Center 和 File Exchange 中查找有关 Reinforcement Learning 的更多信息

标签

产品


版本

R2022b

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by