#ROS导航代码报错：[move_base-3] process has died [pid 7588, exit code 1, cmd /home/wheeltec-client/wheeltec_robot/devel/lib/move_base/move_base __name:=move_base __log:=/home/wheeltec-client/.ros/log/3a3df818-ebe7-11ed-afab-000c2926fa04/move_base-3.log].已解决

导航参考代码

参考这位老哥的RRT*代码：https://zhuanlan.zhihu.com/p/336867657

移植

参考了这位大哥的移植方法：https://blog.csdn.net/bluewhalerobot/article/details/73771020?utm_source=blogxgwz4
拿carrot_planner魔改再一顿复制粘贴就行。理论上来说，只要命名和内容都改完了，就没太大问题了。

报错

Failed to create the rrt_star_planner/RRTSTARPlanner planner, are yousure it is properly registered and that the containing library is built? Exception: MultiLibraryClassLoader: Could not create object of class type rt_star_planner::RRTSTARPlanner as no factory exists for it.Make sure that the library exists and was explicitly loaded through MultiLibraryClassLoader::loadLibrary()

这是一开始发现的报错。结果发现是改代码时 rrt_star_planner::RRTSTARPlanner 写错写成了 rt_star_planner::RRTSTARPlanner。这种错误基本根据报错提示就能改正。
另外，有时也报错：

Failed to create the rrt_star_planner/RRTSTARPlanner planner, are you
sure it is properly registered and that the containing library is built?

的话，貌似可以通过运行setup.bash来使工作环境配置生效。
在工作空间文件夹运行：

source devel/setup.bash

有可能会解决问题。

重难点报错

只针对上述老大哥的代码哈~

process has died [pid 16501, exit code -11, cmd /home/wheeltec-client/wheeltec_robot/devel/lib/move_base/move_base __name:=move_base __log:=/home/wheeltec-client/.ros/log/128979f2-ec12-11ed-afab-000c2926fa04/move_base-3.log].
log file: /home/wheeltec-client/.ros/log/128979f2-ec12-11ed-afab-000c2926fa04/move_base-3*.log

这个报错是在利用Gazebo仿真导航launch文件运行后，再Rviz界面下使用Publish point设置初始点和目标点时会产生的。
由于哥们ROS不太懂，参考ROS官方导航代码planner_core.cpp中有：

bool GlobalPlanner::makePlanService(nav_msgs::GetPlan::Request& req, nav_msgs::GetPlan::Response& resp) {makePlan(req.start, req.goal, resp.plan.poses);resp.plan.header.stamp = ros::Time::now();resp.plan.header.frame_id = frame_id_;return true;
}

这样一个非常神秘的函数，但是老大哥的代码里并没有，于是仿照这个我也写了一个：

bool RRTSTARPlanner::makePlanService(nav_msgs::GetPlan::Request& req, nav_msgs::GetPlan::Response& resp) {makePlan(req.start, req.goal, resp.plan.poses);resp.plan.header.stamp = ros::Time::now();resp.plan.header.frame_id = frame_id_;return true;
}

在初始化函数下还加上了：

make_plan_service_ = private_nh.advertiseService("make_plan", &PlannerWithCostmap::makePlanService, this);

在头文件里加上了函数声明，结果该报错还是报错。（我甚至怀疑是move_base中的代码错了。）

解决

哥们只会笨办法，把 ROS_WARN(“ABCD”); 这个放到makeplan函数里寻找程序到底是在哪个位置崩了。终于发现，在一开始声明节点数组（叫容器吗？我不知道叫什么）时：

std::vector<Node> nodes;

貌似定义了一个空数组，里面没有东西。
接着在寻找最接近的节点时，直接引用这个：

node_nearest = getNearest(nodes, p_rand); // The nearest node of the random point

原函数：

  Node RRTSTARPlanner::getNearest(std::vector<Node> nodes, std::pair<float, float> p_rand){Node node = nodes[0];for (int i = 1; i < nodes.size(); i++){if (distance(nodes[i].x, nodes[i].y, p_rand.first, p_rand.second) < distance(node.x, node.y, p_rand.first, p_rand.second))node = nodes[i];}return node;}

大概是因为其中没有节点，所以就报错了。
于是，在定义了起始节点之后，我就把起始节点加进去了：

Node start_node;start_node.x = start.pose.position.x;start_node.y = start.pose.position.y;start_node.node_id = 0;start_node.parent_id = -1; // None parent nodestart_node.cost = 0.0;nodes.push_back(start_node); //只有这一句是我加的

然后问题就解决了。

总结

我估计就是具体情况具体分析，一般就是某部分代码错了。如果你也可以像我一样，用 ROS_WARN(“ABCD”); 找代码运行到什么位置崩了，那真的是。。。！

ROS利用RRT*导航代码报错process has died [pid 16501, exit code -11的解决方法相关推荐

DS-SLAM的运行[TUM-1] process has died [pid 27902, exit code -11, cmd /home/jerry/catkin_ws/src/DS-SLAM/
心态崩了折腾了一周还是没运行成功上篇博客已经编译通过了,然后下载数据集然后解压,然后别忘了下载tum数据集提供的关联工具工具associate.py 将associate.py拷贝到数据集目录 ...
启动gazebo失败报错[gazebo-1] process has died [pid 10999, exit code 255
启动完Gazebo后,关闭,再次启动出现如下报错: [gazebo-1] process has died [pid 10999, exit code 255, cmd /opt/ros/melodi ...
绝对路径！报错：[gazebo-2] process has died [pid 2382, exit code 134
当执行 roslaunch turtlebot_gazebo turtlebot_world.launch world_file:=/PATH时,报错如下信息: (注:/PATH 是你要打开的worl ...
ubuntu18.04 下运行roscore报错：[rosout-1] process has died [pid 13103, exit code 127
roscore之后报错: [rosout-1] process has died [pid 13103, exit code 127, cmd /opt/ros/melodic/lib/rosout/ ...
[gazebo_gui-2] process has died [pid 4588, exit code 134, cmd /opt/ros/kinetic/lib/gazebo_ros/gzc
当打算算使用gazebo工具导入urdf模型进行仿真,首先在一个终端输入: roscore 再在另一个终端输入: roslaunch pkg_name xxxxx.launch 出现如下错误提 ...
spawner: error: too few arguments [controller_spawner-4] process has died [pid 19013, exit code 2, c
用moveit生成配置文件后,启动roslaunch arm4 demo_gazebo.com报错,如下: spawner: error: too few arguments XmbTextListT ...
[mbot_teleop-1] process has died [pid 3237, exit code 127, cmd /home/arise/sources/catkin_ws/src/mbo
在slam自主导航时候启动键盘控制时候报错错误: [mbot_teleop-1] process has died [pid 3237, exit code 127, cmd /home/aris ...
解决运行gazebo时出现“[gazebo-2] process has died [pid 7920, exit code 255.....“的问题
在运行gazebo时,经常出现"[gazebo-2] process has died [pid 7920, exit code 255, cmd /opt/ros/melodic/lib/ ...
[robot_state_publisher-4] process has died [pid 20533, exit code 1, cmd /opt/ros/melodic/lib/robot_s
sudo gedit /opt/ros/melodic/lib/joint_state_publisher/joint_state_publisher 打开文件后,在文件头加上 reload(sys) ...

ROS利用RRT*导航代码报错process has died [pid 16501, exit code -11的解决方法

导航参考代码

移植

报错

重难点报错

解决

总结

ROS利用RRT*导航代码报错process has died [pid 16501, exit code -11的解决方法相关推荐

最新文章

热门文章