This repository is a planning and evidence-tracking workspace for evaluating whether the Exeed RX can be supported by openpilot.
It is not a working port yet.
- track verified vehicle facts separately from guesses
- collect CAN and UDS evidence before writing port code
- document hardware access, reverse-engineering steps, and blockers
- prepare for a future
opendbcbrand port if the car is actually feasible
- No CAN captures checked in yet
- No ECU firmware query results yet
- No decoded signals yet
- No DBC or
opendbccode yet - Main near-term goal: prove whether lateral control is accessible on a bus that can be observed and replayed safely
| File | Purpose |
|---|---|
| PLAN.md | Execution plan with decision gates and deliverables |
| RESEARCH.md | Verified facts, hypotheses, unknowns, and current architecture notes |
| HARDWARE.md | Passive sniffing and later interception hardware guidance |
| CAN_NOTES.md | Evidence log for captures, firmware responses, and candidate signals |
| COMPARISON_NOTES.md | Cross-checks against Jaecoo J7, Exeed VX, and other external references |
| FIRST_DRIVE_CHECKLIST.md | Conservative first logging workflow for the initial passive session |
| captures/ | Naming convention and storage guidance for raw logs and sidecar notes |
| history/ | Imported conversations and external research snapshots kept for provenance |
Imported conversations in history/ are reference inputs, not primary evidence. Keep anything copied from them labeled as a lead, comparison point, or hypothesis until verified on the actual car.
- Verified vehicle year, trim, and ADAS feature set
- At least one bus capture from OBD and, if needed, camera or ADAS wiring
- ECU firmware query results saved in the repo
- Candidate steer, speed, and ACC-state signals identified from real data
- Initial
opendbclayout drafted based on current upstream structure
Complete passive sniffing and determine whether the stock LKAS path is visible on OBD, on a camera harness bus, or not visible at all.