Example Code for AT command
Last revision 2026/06/08
This article compiles AT command example codes and UART AT communication, covering AT+INVOKE inference calls, model switching AT+MODEL, TSCORE/TIOU/TSIMILARITY threshold configuration, and gesture/posture learning instructions. Guidance path: Docs → Tutorials → Projects → SKU → Category.
AT Command Tutorial
Common AT commands are listed below. For more, please refer to the additional documentation: SEN0670_UART_Communication_Protocol_V.10.pdf.
| AT Command | Description | Response |
|---|---|---|
AT+ID? |
Query device ID | {"type":0,"name":"ID?","code":0,"code_text":"EL_OK","ok":1,"data":"3343029e"} |
AT+NAME? |
Query device name | {"type":0,"name":"NAME?","code":0,"code_text":"EL_OK","ok":1,"data":"DFRobot Human Pose"} |
AT+VER? |
Query firmware version | {"type": 0, "name": "VER?", "code": 0, "code_text": "EL_OK", "ok": 1, "data": {"at_api": "v0", "software": "2026.04.14", "hardware": "v1.0.0"}} |
AT+RST |
Restart device | The device resets immediately; usually no complete success packet is received. |
AT+MODELS? |
Query model list | {"type": 0, "name": "MODELS?", "code": 0, "code_text": "EL_OK", "ok": 1, "data": [{"id": 4, "type": 0, "address": 11669504, "size": 3010560}, {"id": 3, "type": 0, "address": 9322496, "size": 2347008}, {"id": 2, "type": 0, "address": 6963200, "size": 2359296}, {"id": 1, "type": 0, "address": 4194304, "size": 2768896}]} |
AT+MODEL? |
Query current model | {"type": 0, "name": "MODEL?", "code": 0, "code_text": "EL_OK", "ok": 1, "data": {"id": 3, "type": 0, "address": 9322496, "size": 2347008}} |
AT+MODEL=1 |
Switch to hand model (custom gesture) | {"type": 0, "name": "MODEL", "code": 0, "code_text": "EL_OK", "ok": 1, "data": {"model": {"id": 1, "type": 0, "address": 4194304, "size": 2768896}}} |
AT+MODEL=3 |
Switch to posture model (custom posture) | {"type": 0, "name": "MODEL", "code": 0, "code_text": "EL_OK", "ok": 1, "data": {"model": {"id": 3, "type": 0, "address": 9322496, "size": 2347008}}} |
AT+MODEL=4 |
Switch to gesture model (fixed gestures) | {"type": 0, "name": "MODEL", "code": 0, "code_text": "EL_OK", "ok": 1, "data": {"model": {"id": 4, "type": 0, "address": 11669504, "size": 3010560}}} |
AT+INVOKE=<n_times>,<differed>,<result_only> |
Perform inference, e.g., one inference only result: AT+INVOKE=1,0,1 |
{"type": 0, "name": "INVOKE", "code": 0, "data": {"timestamp_ms": 1248979, "model": {"id": 4, "type": 0, "address": 11669504, "size": 3010560}, "algorithm": {"type": 6, "categroy": 1, "input_from": 1, "config": {"tscore": 60, "tiou": 45}}, "sensor": {"id": 1, "type": 1, "state": 1, "opt_id": 0, "opt_detail": "240x240 Auto"}}} {"type": 1, "name": "INVOKE", "code": 0, "data": {"count": 1, "seq": 1, "timestamp_ms": 1249074, "image_base64_len": 0, "perf": [9, 85, 0], "boxes": [], "resolution": [240, 240]}} |
AT+BREAK |
Stop current task | {"type": 0, "name": "BREAK", "code": 0, "code_text": "EL_OK", "ok": 1, "data": 1346635} |
AT+TPROTO=0 |
Switch to JSON output | {"type": 0, "name": "TPROTO", "code": 0, "data": 0, "message": "TPROTO updated"} |
AT+TPROTO? |
Query output protocol | {"type":0,"name":"TPROTO?","code":0,"data":0} |
AT+TKPTS=0 |
Disable keypoint output | {"type": 0, "name": "TKPTS", "code": 0, "data": 0, "message": "TKPTS updated"} |
AT+TKPTS=1 |
Enable keypoint output | {"type": 0, "name": "TKPTS", "code": 0, "data": 1, "message": "TKPTS updated"} |
AT+TKPTS? |
Query keypoint output configuration | {"type": 0, "name": "TKPTS?", "code": 0, "data": 1} |
AT+TSCORE? |
Query confidence threshold | {"type": 0, "name": "TSCORE?", "code": 0, "code_text": "EL_OK", "ok": 1, "data": 60} |
AT+TSCORE=60 |
Set confidence threshold to 60 | {"type":0,"name":"TSCORE","code":0,"code_text":"EL_OK","ok":1,"data":60} |
AT+TIOU? |
Query IoU threshold | {"type": 0, "name": "TIOU?", "code": 0, "code_text": "EL_OK", "ok": 1, "data": 45} |
AT+TIOU=45 |
Set IoU threshold to 45 | {"type": 0, "name": "TIOU", "code": 0, "code_text": "EL_OK", "ok": 1, "data": 45} |
AT+TSIMILARITY? |
Query learning similarity threshold | {"type": 0, "name": "TSIMILARITY?", "code": 0, "code_text": "EL_OK", "ok": 1, "data": 60} |
AT+TSIMILARITY=80 |
Set learning similarity threshold to 80 | {"type": 0, "name": "TSIMILARITY", "code": 0, "code_text": "EL_OK", "ok": 1, "data": 80} |
AT+HANDAUTOLEARN |
Automatically name and start gesture learning | {"type": 0, "name": "HANDAUTOLEARN", "code": 0, "data": {"hand_name": "Gesture_1"}} |
AT+HANDLIST? |
Query learned gestures | {"type": 0, "name": "HANDLIST?", "code": 0, "data": []} |
AT+HANDCLEAR |
Clear all gestures | {"type": 0, "name": "HANDCLEAR", "code": 0} |
AT+POSEAUTOLEARN |
Automatically name and start posture learning | {"type": 0, "name": "POSEAUTOLEARN", "code": 0, "data": {"pose_name": "Posture_1"}} |
AT+POSELIST? |
Query learned postures | {"type": 0, "name": "POSELIST?", "code": 0, "data": []} |
AT+POSECLEAR |
Clear all postures | {"type": 0, "name": "POSECLEAR", "code": 0} |
AT Communication Example
Preparation: Connect the sensor to the PC via USB and open a serial terminal software such as sscom.
Was this article helpful?