Prevent optimization thread from doing its work when a stop was requested (#391)

Author: carlos-m159

fuse_optimizers/src/batch_optimizer.cpp

@@ -157,6 +157,11 @@ void BatchOptimizer::optimizationLoop()
     }
     {
       std::lock_guard<std::mutex> lock(optimization_mutex_);
+      if (!started_)
+      {
+        ROS_DEBUG_STREAM("Optimizer stopped while trying to acquire optimization_mutex_. Skipping optimization.");
+        continue;
+      }
       // Copy the combined transaction so it can be shared with all the plugins
       fuse_core::Transaction::ConstSharedPtr const_transaction;
       {
@@ -289,19 +294,19 @@ void BatchOptimizer::start()
 void BatchOptimizer::stop()
 {
   ROS_INFO_STREAM("Stopping optimizer.");
-  // Tell all the plugins to stop
-  stopPlugins();
+  started_ = false;
   // Reset the optimizer state
   {
     std::lock_guard<std::mutex> lock(optimization_requested_mutex_);
     optimization_request_ = false;
   }
-  started_ = false;
   // DANGER: The optimizationLoop() function obtains the lock optimization_mutex_ lock and the
   //         combined_transaction_mutex_ lock at the same time. We perform a parallel locking scheme here to
   //         prevent the possibility of deadlocks.
   {
     std::lock_guard<std::mutex> lock(optimization_mutex_);
+    // Tell all the plugins to stop
+    stopPlugins();
     // Clear the combined transation
     {
       std::lock_guard<std::mutex> lock(combined_transaction_mutex_);

fuse_optimizers/src/fixed_lag_smoother.cpp

@@ -210,6 +210,13 @@ void FixedLagSmoother::optimizationLoop()
     // Optimize
     {
       std::lock_guard<std::mutex> lock(optimization_mutex_);
+      // Make sure stop was not called while we were trying to acquire optimization mutex
+      if (!started_)
+      {
+        ROS_DEBUG_STREAM("Optimizer stopped while trying to acquire optimization_mutex_. Skipping optimization.");
+        continue;
+      }
+
       // Apply motion models
       auto new_transaction = fuse_core::Transaction::make_shared();
       // DANGER: processQueue obtains a lock from the pending_transactions_mutex_
@@ -492,21 +499,21 @@ void FixedLagSmoother::start()
 void FixedLagSmoother::stop()
 {
   ROS_INFO_STREAM("Stopping optimizer.");
-  // Tell all the plugins to stop
-  stopPlugins();
+  started_ = false;
+  ignited_ = false;
   // Reset the optimizer state
   {
     std::lock_guard<std::mutex> lock(optimization_requested_mutex_);
     optimization_request_ = false;
   }
-  started_ = false;
-  ignited_ = false;
   setStartTime(ros::Time(0, 0));
   // DANGER: The optimizationLoop() function obtains the lock optimization_mutex_ lock and the
   //         pending_transactions_mutex_ lock at the same time. We perform a parallel locking scheme here to
   //         prevent the possibility of deadlocks.
   {
     std::lock_guard<std::mutex> lock(optimization_mutex_);
+    // Tell all the plugins to stop
+    stopPlugins();
     // Clear all pending transactions
     {
       std::lock_guard<std::mutex> lock(pending_transactions_mutex_);