Skip to content

perception-test - #15

Merged
SuBamm88 merged 1 commit into
mainfrom
perception-branch
Jun 1, 2026
Merged

perception-test#15
SuBamm88 merged 1 commit into
mainfrom
perception-branch

Conversation

@SuBamm88

@SuBamm88 SuBamm88 commented Jun 1, 2026

Copy link
Copy Markdown
Owner

No description provided.

@SuBamm88
SuBamm88 merged commit 590ebb4 into main Jun 1, 2026
1 check passed

@chatgpt-codex-connector chatgpt-codex-connector Bot left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 741e6fb204

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

executable="cctv_to_map",
name="cctv_to_map",
output="screen",
parameters=[cctv_to_map_params],

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P1 Badge Avoid dual writers on the costmap topic

In perception_demo.launch.py this starts cctv_to_map with config/cctv_to_map.yaml, whose objects_topic is /cctv/objects_map, while the fusion node also hard-codes a publisher on /cctv/objects_map in fusion_node.py. When this demo launch is used, the Nav2 layer subscribes to alternating raw CCTV PoseArrays and fused PoseArrays, so LiDAR-only or fused obstacles can disappear whenever the raw CCTV publisher ticks. Point the raw cctv_to_map PoseArray output at an unused topic in this launch, or make fusion the only publisher to /cctv/objects_map.

Useful? React with 👍 / 👎.

# 매칭 안 된 CCTV 객체(LiDAR 사각지대) → ID 충돌 방지 오프셋 적용
for j, c in enumerate(cctv):
if j not in matched_cctv:
c.id = c.id + CCTV_ID_OFFSET

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P2 Badge Avoid repeatedly offsetting cached CCTV IDs

When an unmatched CCTV object is fused, this mutates the same TrackedObject instance stored in self.cctv_by_topic; because _fuse_and_publish runs on a timer and can execute multiple times before the next CCTV callback, the same object gets CCTV_ID_OFFSET added every cycle. With the default 10 Hz publish rate and 1 s CCTV timeout, one detection can be published as 100005, 200005, etc., breaking the stable-ID contract and potentially overflowing the int32 message field. Build a copied output object or derive the offset ID without modifying the cached input.

Useful? React with 👍 / 👎.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant