fix: clone actions_scaled tensor for output calculations

.pre-commit-config.yaml

@@ -22,4 +22,4 @@ repos:
           - --comment-style
           - /*| *| */
 
-exclude: ^(src/rl_sar/library/matplotlibcpp|src/rl_sar/library/unitree_legged_sdk_3.2|src/rl_sar/library/unitree_sdk2)/
+exclude: ^(src/rl_sar/library/core/matplotlibcpp|src/rl_sar/library/thirdparty)/

src/rl_sar/library/core/rl_sdk/rl_sdk.cpp

@@ -103,12 +103,12 @@ void RL::InitControl()
 void RL::ComputeOutput(const torch::Tensor &actions, torch::Tensor &output_dof_pos, torch::Tensor &output_dof_vel, torch::Tensor &output_dof_tau)
 {
     torch::Tensor actions_scaled = actions * this->params.action_scale;
-    torch::Tensor pos_actions_scaled = actions_scaled;
+    torch::Tensor pos_actions_scaled = actions_scaled.clone();
     torch::Tensor vel_actions_scaled = torch::zeros_like(actions);
     for (int i : this->params.wheel_indices)
     {
         pos_actions_scaled[0][i] = 0.0;
-        vel_actions_scaled[0][i] = actions[0][i];
+        vel_actions_scaled[0][i] = actions_scaled[0][i];
     }
     torch::Tensor all_actions_scaled = pos_actions_scaled + vel_actions_scaled;
     output_dof_pos = pos_actions_scaled + this->params.default_dof_pos;