Can JetRacer_ROS Kit run yolo and DQN simultanteously?

I bought JetRacer ROS KIT.
JetRacer use JETSON NANO B01 and Rasberi PI 2090 controller.

I want to run two network at the same time.
So, Can Jetracer run two network at a time?

A.
Yolo → JetRacer → Run

B.
DQN → JetRacer → run

C.
Yolo and DQN → JetRacer → Run

Not A and B case, I want C case.
Yolo does not matter which version.

DQN network is
class Qnet(nn.Module):

def init(self):
super(Qnet, self).init()

self.fc1 = nn.Linear(4, 128) # 4 input states
self.fc2 = nn.Linear(128, 128)
self.fc3 = nn.Linear(128, 128)
self.fc4 = nn.Linear(128, 128)
self.fc5 = nn.Linear(128, 7)

def forward(self, x):

x = F.relu(self.fc1(x))
x = F.relu(self.fc2(x))
x = F.relu(self.fc3(x))
x = F.relu(self.fc4(x))

x = self. fc5(x)

return x

Hi @dlwogns0207, seeing as JetRacer doesn’t support these out-of-the-box, I’m guessing you mean if you were to add them - in which case, is the DQN reinforcement learning model doing training at runtime, or inference only? Due to the simplicitly of the Qnet model you’ve defined, I think inference should be okay there in combination with YOLO (albeit at a reduced framerate perhaps). You can find some YOLOv3 benchmarks for Jetson Nano in this post:


I would start by attempting to run both the YOLO model you want and your DQN model at the same on your Nano and see what the performance is.

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.