feat: Handle situations that do not require clamp

src/rl_sar/library/rl_sdk/rl_sdk.cpp

@@ -397,8 +397,16 @@ void RL::ReadYaml(std::string robot_name)
     this->params.num_observations = config["num_observations"].as<int>();
     this->params.observations = ReadVectorFromYaml<std::string>(config["observations"]);
     this->params.clip_obs = config["clip_obs"].as<double>();
-    this->params.clip_actions_upper = torch::tensor(ReadVectorFromYaml<double>(config["clip_actions_upper"], this->params.framework, rows, cols)).view({1, -1});
-    this->params.clip_actions_lower = torch::tensor(ReadVectorFromYaml<double>(config["clip_actions_lower"], this->params.framework, rows, cols)).view({1, -1});
+    if(config["clip_actions_lower"] && config["clip_actions_upper"])
+    {
+        this->params.clip_actions_upper = torch::tensor(ReadVectorFromYaml<double>(config["clip_actions_upper"], this->params.framework, rows, cols)).view({1, -1});
+        this->params.clip_actions_lower = torch::tensor(ReadVectorFromYaml<double>(config["clip_actions_lower"], this->params.framework, rows, cols)).view({1, -1});
+    }
+    else
+    {
+        this->params.clip_actions_upper = torch::tensor({}).view({1, -1});
+        this->params.clip_actions_lower = torch::tensor({}).view({1, -1});
+    }
     this->params.action_scale = config["action_scale"].as<double>();
     this->params.hip_scale_reduction = config["hip_scale_reduction"].as<double>();
     this->params.hip_scale_reduction_indices = ReadVectorFromYaml<int>(config["hip_scale_reduction_indices"]);

src/rl_sar/scripts/rl_sdk.py

@@ -387,8 +387,12 @@ def ReadYaml(self, robot_name):
         self.params.action_scale = config["action_scale"]
         self.params.hip_scale_reduction = config["hip_scale_reduction"]
         self.params.hip_scale_reduction_indices = config["hip_scale_reduction_indices"]
-        self.params.clip_actions_upper = torch.tensor(self.ReadVectorFromYaml(config["clip_actions_upper"], self.params.framework, rows, cols)).view(1, -1)
-        self.params.clip_actions_lower = torch.tensor(self.ReadVectorFromYaml(config["clip_actions_lower"], self.params.framework, rows, cols)).view(1, -1)
+        if config["clip_actions_upper"] and config["clip_actions_upper"]:
+            self.params.clip_actions_upper = torch.tensor(self.ReadVectorFromYaml(config["clip_actions_upper"], self.params.framework, rows, cols)).view(1, -1)
+            self.params.clip_actions_lower = torch.tensor(self.ReadVectorFromYaml(config["clip_actions_lower"], self.params.framework, rows, cols)).view(1, -1)
+        else:
+            self.params.clip_actions_upper = None
+            self.params.clip_actions_lower = None
         self.params.num_of_dofs = config["num_of_dofs"]
         self.params.lin_vel_scale = config["lin_vel_scale"]
         self.params.ang_vel_scale = config["ang_vel_scale"]

src/rl_sar/scripts/rl_sim.py

@@ -208,8 +208,10 @@ def Forward(self):
             actions = self.model.forward(history_obs)
         else:
             actions = self.model.forward(clamped_obs)
-        clamped_actions = torch.clamp(actions, self.params.clip_actions_lower, self.params.clip_actions_upper)
-        return clamped_actions
+        if self.params.clip_actions_lower is not None and self.params.clip_actions_upper is not None:
+            return torch.clamp(actions, self.params.clip_actions_lower, self.params.clip_actions_upper)
+        else:
+            return actions
 
     def ThreadControl(self):
         thread_period = self.params.dt

src/rl_sar/src/rl_real_a1.cpp

@@ -181,9 +181,14 @@ torch::Tensor RL_Real::Forward()
 
     torch::Tensor actions = this->model.forward({this->history_obs}).toTensor();
 
-    torch::Tensor clamped_actions = torch::clamp(actions, this->params.clip_actions_lower, this->params.clip_actions_upper);
-
-    return clamped_actions;
+    if(this->params.clip_actions_upper.numel() != 0 && this->params.clip_actions_lower.numel() != 0)
+    {
+        return torch::clamp(actions, this->params.clip_actions_lower, this->params.clip_actions_upper);
+    }
+    else
+    {
+        return actions;
+    }
 }
 
 void RL_Real::Plot()

src/rl_sar/src/rl_sim.cpp

@@ -259,8 +259,14 @@ torch::Tensor RL_Sim::Forward()
         actions = this->model.forward({clamped_obs}).toTensor();
     }
 
-    torch::Tensor clamped_actions = torch::clamp(actions, this->params.clip_actions_lower, this->params.clip_actions_upper);
-    return clamped_actions;
+    if(this->params.clip_actions_upper.numel() != 0 && this->params.clip_actions_lower.numel() != 0)
+    {
+        return torch::clamp(actions, this->params.clip_actions_lower, this->params.clip_actions_upper);
+    }
+    else
+    {
+        return actions;
+    }
 }
 
 void RL_Sim::Plot()