Pixkit Autonomous Driving Development Kit

Pixmoving Inc

Who Are We

PIX Moving is forged by an interdisciplinary and innovative team of around 200 members from diverse cultural backgrounds with in-depth expertise in various fields. Software engineer, algorithm engineer, electronic engineer, designer, architect, entrepreneur and artist from 7 countries including Italy, the US, Pakistan, India and Japan are bound together with the same vision to build the bridge between future mobility and digital manufacturing. Currently PIX Moving has got fundings from various investors around the globe including VC SOSV, A-listed company Guizhou Transportation Planning Survey&design Academe Co.,ltd. and TIS from Japan.

PIXKIT

pixkit-architecture

Provide a one-stop autonomous driving development solution, offering customers a 0-1 autonomous driving development platform.
Complete the integration of computing platform, wire-controlled chassis, sensors, and autonomous driving systems for customers, allowing them to focus on algorithm development.
Provide mapping and sensor calibration toolchains to reduce the development threshold.
Open-source all source code and provide detailed documentation to make development easier.
Continuously update and iterate software and algorithms to stay at the forefront of technology.
Benefit from the resources of ROS/Autoware community, providing continuous motivation for development.
Provide different ODD examples to expand Autoware to different scenarios (support to be provided in the future).

Hardware component

1.Electric Open-source Pix Ultra-Skateboard DBW Chassis

pix-chassis

2.Sensor Package And Modular Sensor Brancket

pixkit-sensors

3.Industrial Remote Controller For The Chassis

remote-controller

Tool Chains for Autonomous Driving Development

Sensor Calibration

camera Intrisics calibration

image_width: 1920
image_height: 1080
camera_name: trafficlight
camera_matrix:
  rows: 3
  cols: 3
  data: [1010.2744 ,    0.     ,  941.65588,
            0.     , 1008.70001,  517.61131,
            0.     ,    0.     ,    1.     ]
distortion_model: plumb_bob
distortion_coefficients:
  rows: 1
  cols: 5
  data: [-0.336502, 0.091380, -0.001975, 0.000444, 0.000000]
rectification_matrix:
  rows: 3
  cols: 3
  data: [1., 0., 0.,
         0., 1., 0.,
         0., 0., 1.]
projection_matrix:
  rows: 3
  cols: 4
  data: [723.01111,   0.     , 947.26543,   0.     ,
           0.     , 895.51398, 508.08632,   0.     ,
           0.     ,   0.     ,   1.     ,   0.     ]

LiDAR-camera calibration

rs162camera:
  x: 0.5789871
  y: -0.08247789
  z: -0.3453706
  roll: -1.55877693
  pitch: 0.00713767
  yaw: -1.55820481

camera-lidar-calibration - IMU calibration

Gyr:
    unit: " rad/s"
    avg-axis:
    gyr_n: 1.5059072284923697e-03
    gyr_w: 4.3430855283551206e-05
    x-axis:
    gyr_n: 1.6901233770452774e-03
    gyr_w: 5.0850707578827144e-05
    y-axis:
    gyr_n: 1.3392742394140514e-03
    gyr_w: 3.7654685426892668e-05
    z-axis:
    gyr_n: 1.4883240690177805e-03
    gyr_w: 4.1787172844933785e-05
Acc:
    unit: " m/s^2"
    avg-axis:
    acc_n: 5.9215155351791055e-03
    acc_w: 1.3379378640306186e-04
    x-axis:
    acc_n: 6.0017230453598448e-03
    acc_w: 1.0726720420556991e-04
    y-axis:
    acc_n: 6.7689914243794181e-03
    acc_w: 1.6961241589651517e-04
    z-axis:
    acc_n: 4.9938321357980535e-03
    acc_w: 1.2450173910710051e-04