nav tooling

Recording Mid360 + Go2 run

TODO

Testing KronNav on a dataset

build kronknav

uv run maturin develop --uv --release -m dimos/mapping/ray_tracing/rust/Cargo.toml && uv run maturin develop --uv --release -m dimos/navigation/nav_3d/mls_planner/rust/Cargo.toml

run on a dataset

uv run python -m dimos.navigation.nav_3d.mls_planner.utils.plan_rrd ~/recordings/china_office_pointlio_sensor.db --robot-height 0.4 --lidar-stream pointlio_lidar --odom-stream pointlio_odom

How to Optimize Point-LIO Configs

1. Record mid360 with PCAP enabled, see the go2_mid360.md doc (in dimos)
2. Use pcap_to_db.py to generate alternative lidar/odom outcomes - renders to rerun

Pcap to DB

You can see/edit the default Point-LIO config here.

python -m dimos.hardware.sensors.lidar.pointlio.scripts.pcap_to_db --pcap mid360_shake_stairs/mid360_shake_stairs.pcap --db stairs.db

will open up rerun with aggregated map + odom path create a new db file with the pointlio stream

generate a global map from this via

dimos map global --lidar pointlio_lidar --pgo-tol=0 --no-carve

MacOS Caveats

The module aliases the synthetic IPs onto lo0, which needs sudo. A tty-less worker can't prompt, so set up the interface by hand, then pass --no-network-setup:

sudo ifconfig lo0 alias 192.168.1.5 netmask 255.255.255.0
sudo ifconfig lo0 alias 192.168.1.155 netmask 255.255.255.0
sudo route -n add -host 224.1.1.5 -interface lo0
sudo route -n add -host 255.255.255.255 -interface lo0

python -m dimos.hardware.sensors.lidar.pointlio.scripts.pcap_to_db \
    --pcap "$PCAP" --no-network-setup