本文整理汇总了C++中Simulator::close方法的典型用法代码示例。如果您正苦于以下问题：C++ Simulator::close方法的具体用法？C++ Simulator::close怎么用？C++ Simulator::close使用的例子？那么, 这里精选的方法代码示例或许可以为您提供帮助。您也可以进一步了解该方法所在类Simulator的用法示例。
在下文中一共展示了Simulator::close方法的1个代码示例，这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞，您的评价将有助于系统推荐出更棒的C++代码示例。
示例1: main
int main(int argc, char *argv[]) {
// std::mt19937 random_source(0);
std::minstd_rand random_source(0);
std::uniform_real_distribution<float> angledist(-M_PI / 4, M_PI / 4);
Simulator *sim = new Simulator();
// Sets resolution. Default is 320X240
sim->setCameraResolution(640,480);
// Initialize the simulator. Further camera configuration won't take any effect from now on.
sim->init();
// Run this many episodes
int episodes = 10;
for (int i = 0; i < episodes; ++i) {
// Starts a new episode. It is not needed right after init() but it doesn't cost much and the loop is nicer.
sim->newEpisode("2t7WUuJeko7"); // Take optional viewpoint_id argument, otherwise launches at a random location
for (int step = 0;step < 10;++step) {
// Get the state
SimStatePtr state = sim->getState(); // SimStatePtr is std::shared_ptr<SimState>
// Which consists of:
unsigned int n = state->step;
cv::Mat rgb = state->rgb; // Use OpenCV CV_8UC3 type (i.e. 8bit color rgb)
cv::Mat depth = state->depth; // ignore for now
ViewpointPtr location = state->location; // Need a class to hold viewpoint id, and cv::Point3 for x,y,z location of a viewpoint
float heading = state->heading;
float elevation = state->elevation; // camera parameters
std::vector<ViewpointPtr> reachable = state->navigableLocations; // Where we can move to,
std::uniform_int_distribution<> distribution(0, reachable.size() - 1);
int locationIdx = distribution(random_source);
float headingChange = angledist(random_source);
float elevationChange = angledist(random_source);
/* std::stringstream ss;
ss << "rgb_" << i << "_" << step << ".png";
cv::imwrite(ss.str(), rgb);*/
// Make action (index into reachable, heading change in rad, elevation change in rad)
// E.g. an RL agent will sample an action here. A reward can be determined based on location, heading, elevation but that is dataset dependent
sim->makeAction(locationIdx, headingChange, elevationChange);
std::this_thread::sleep_for(std::chrono::milliseconds{100});
}
}
// It will be done automatically in destructor but after close You can init it again with different settings.
sim->close();
delete sim;
return 0;
}