main content

create custom turn-凯发k8网页登录

create custom turn-based multiagent reinforcement learning environment

since r2023b

description

use rlturnbasedfunctionenv to create a custom turn-based multiagent reinforcement learning environment in which agents execute in turns. to create your custom environment, you supply the observation and action specifications as well as your own reset and step matlab® functions. to verify the operation of your environment, rlturnbasedfunctionenv automatically calls validateenvironment after creating the environment.

creation

syntax

env = rlturnbasedfunctionenv(observationinfo,actioninfo,stepfcn,resetfcn)

description

example

env = rlturnbasedfunctionenv(observationinfo,actioninfo,stepfcn,resetfcn) creates a turn-based multiagent environment using observation and action specifications and custom step and reset functions. the cell arrays observationinfo and actioninfo must contain the observation and action specifications, respectively, for each agent. the stepfcn and resetfcn arguments are the names of your step and reset matlab functions, respectively, and they are used to set the stepfcn and resetfcn properties of env.

input arguments

observation specifications, specified as a cell array with as many elements as the number of agents. every element of the cell must contain the observation specifications for a corresponding agent. the observation specification for an agent must be an rlfinitesetspec or rlnumericspec object or a vector containing a mix of such objects (in which case every element of the vector defines the properties of a specific observation channel for the agent).

action specifications, specified as a cell array with as many elements as the number of agents. every element of the cell must contain the observation specifications for a corresponding agent. the action specification for an agent must be an rlfinitesetspec (for discrete action spaces) or rlnumericspec (for continuous action spaces) object. this object defines the properties of the action channel for the agent.

note

only one action channel per agent is allowed.

properties

environment step function, specified as a function name, function handle, or handle to an anonymous function. the sim and train functions call stepfcn to update the environment at every simulation or training step.

this function must have two inputs and four outputs, as illustrated by the following signature.

[nextobservation,reward,isdone,updatedinfo] = mystepfunction(action,info)

for a given action input, the step function returns the values of the next observation and reward, a logical value indicating whether the episode is terminated, and an updated environment information variable.

specifically, the required input and output arguments are:

  • action — cell array containing the current actions from the agents that are currently executing. must contain as many elements as the number of agents that are executing at the current step. each element of the action cell must match the dimensions and data type specified in the corresponding element of the actioninfo(activeagentindex) cell.

  • info and updatedinfo — this must be a structure containing the field activeagentindex, which is a scalar or vector of indices indicating the agents that are active in the current step. the environment step function can modify the value to control the execution of agents in the next step. other optional fields can contain the environment state and parameters or any data that you want to pass from one step to the next.

    the simulation or training functions (train or sim) handle this variable by:

    1. initializing info using the second output argument returned by resetfcn, at the beginning of the episode

    2. passing info as second input argument to stepfcn at each training or simulation step

    3. updating info using the fourth output argument returned by stepfcn, updatedinfo

  • nextobservation — cell array containing the next observations for all the agents. these are the observations related to the next state (the transition to the next state is caused by the current actions contained in action). therefore, nextobservation must contain as many elements as the number of agents and each element must match the dimensions and data types specified in the corresponding element of the observationinfo cell.

  • reward — vector containing the rewards for all the agents. these are the rewards generated by the transition from the current state to the next one. each element of the vector must be a numeric scalar.

  • isdone — logical value indicating whether to end the simulation or training episode.

to use additional input arguments beyond the allowed two, define your additional arguments in the matlab workspace, then specify stepfcn as an anonymous function that in turn calls your custom function with the additional arguments defined in the workspace, as shown in the example create custom environment using step and reset functions.

example: stepfcn="mystepfcn"

environment reset function, specified as a function name, function handle, or handle to an anonymous function. the sim function calls your reset function to reset the environment at the start of each simulation, and the train function calls it at the start of each training episode.

the reset function that you provide must have no inputs and two outputs, as illustrated by the following signature.

[initialobservation,info] = myresetfunction()

the reset function sets the environment to an initial state and computes the initial value of the observation. for example, you can create a reset function that randomizes certain state values such that each training episode begins from different initial conditions. the initialobservation must be a cell array containing the initial observations for all the agents. therefore, initialobservation must contain as many elements as the number of agents and each element must match the dimensions and data types specified in the corresponding element of the observationinfo cell.

the info output of resetfcn initializes the info property of your environment and contains any data that you want to pass from one step to the next. this can be the environment state or a structure containing state and parameters. the simulation or training function (train or sim) supplies the current value of info as the second input argument of stepfcn, then uses the fourth output argument returned by stepfcn to update the value of info.

to use additional input arguments beyond the allowed two, define your argument in the matlab workspace, then specify stepfcn as an anonymous function that in turn calls your custom function with the additional arguments defined in the workspace, as shown in the example create custom environment using step and reset functions.

example: resetfcn="myresetfcn"

information to pass to next step, specified as a structure containing the field activeagentindex, which is a scalar or vector of indexes indicating the agents that are active in the current step. the environment step function can modify this field to control the execution of agents in the next step. other optional fields can contain the environment state and parameters or any data that you want to pass from one step to the next.

when resetfcn is called, whatever you define as the info output of resetfcn initializes this property. when a step occurs the simulation or training function (train or sim) uses the current value of info as the second input argument for stepfcn. once stepfcn completes, the simulation or training function then updates the current value of info using the fourth output argument returned by stepfcn.

example: info.activeagentindex=[2 3]

object functions

getactioninfoobtain action data specifications from reinforcement learning environment, agent, or experience buffer
getobservationinfoobtain observation data specifications from reinforcement learning environment, agent, or experience buffer
traintrain reinforcement learning agents within a specified environment
simsimulate trained reinforcement learning agents within specified environment
validateenvironmentvalidate custom reinforcement learning environment

examples

create a custom turn-based multiagent environment by supplying custom matlab® functions. using rlturnbasedfunctionenv, you can create a custom matlab reinforcement learning environment in which agents execute in turns. to create your custom turn-based environment, you must define observation specifications, action specifications, and step and reset functions.

for this example, consider an environment containing four agents, all of them having a continuous observation space, and receiving observation vectors of four, two, five, and three elements respectively.

define the agent observation spaces using a cell array.

obsinfo = { rlnumericspec([4 1]), ...
            rlnumericspec([2 1]), ...
            rlnumericspec([5 1]), ...
            rlnumericspec([3 1]) };

for this example, the first and fourth agents have a finite action set containing two and four elements, respectively, while the second and the third have continuous action spaces consisting of a scalar and a two-dimensional vector. define the agent action sets and spaces using a cell array.

actinfo = { rlfinitesetspec([1 2]), ...
            rlnumericspec([1 1]), ...
            rlnumericspec([2 1]), ...
            rlfinitesetspec([1 2 3 4]) };

next, specify your step and reset functions. for this example, use the functions resetfcn and stepfcn defined at the end of the example.

note that while the custom reset and step functions that you must pass to rlturnbasedfuntionenv must have exactly zero and two arguments, respectively, you can avoid this limitation by using anonymous functions. for an example on how to do this, see create custom environment using step and reset functions.

to create the custom turn-based multiagent function environment, use rlturnbasedfunctionenv.

env = rlturnbasedfunctionenv( ...
    obsinfo,actinfo, ...
    @stepfcn,@resetfcn)
env = 
  rlturnbasedfunctionenv with properties:
     stepfcn: @stepfcn
    resetfcn: @resetfcn
        info: [1x1 struct]

you can now create agents for env and train or simulate them as you would for any other environment.

environment functions

environment reset function.

function [initialobs, info] = resetfcn()
% resetfun sets the default state of the environment.
%
% - initialobs is a 1xn cell array.
% - info is a structure with the field:
%     - activeagentindex: a scalar or vector of agent indices that
%       are active in the current step. the value can be modified to
%       control the execution of agents in the next step.
%     - other fields of any matlab data type can be used to pass
%       information from one step to the next.
% for this example, initialize the agent observations randomly 
initialobs = {
    rand([4 1]), ...
    rand([2 1]), ...
    rand([5 1]), ...
    rand([3 1]) };
% initialize the info structure
info.environmentstate = initialobs;
info.executionorder = {1, [2,3], 4};
info.turncount = 1;
info.activeagentindex = 1;
end

environment step function.

function [nextobs, reward, isdone, info] = stepfcn(action, info)
% stepfun specifies how the environment advances to the next state 
% given the actions from all the agents.
% 
% - nextobs is a 1xn cell array (n is the total number of agents).
% - action is a 1xp cell array (p is the number of active agents).
% - reward is a 1xn numeric array.
% - isdone is a logical or numeric scalar.
% - info is a structure with the field:
%     - activeagentindex: a scalar or vector of agent indices that
%       are active in the current step. the value can be modified to
%       control the execution of agents in the next step.
%     - other fields of any matlab data type can be used to pass
%       information from one step to the next.
% for this example, just return to each agent a random observation 
nextobs = {
    rand([4 1]), ...
    rand([2 1]), ...
    rand([5 1]), ...
    rand([3 1]) };
% return a random reward vector multiplied by the norm of the action
% of the first or the current executing agents
reward = rand(4,1)*norm(action{1});
% return a false is-done value.
isdone = false;
% extract the execution order and turn count
ord = info.executionorder;
tc  = info.turncount;
% reset turn count to zero when it reaches 3
if mod(tc, numel(ord)) == 0
    tc = 0;
end
% set activeagentindex and turncount fields
info.activeagentindex = ord{tc 1};
info.turncount = tc 1;
% set the environmentstate field
info.environmentstate = nextobs;
end

version history

introduced in r2023b

see also

functions

objects

topics

网站地图