Implementation of Autonomous Control Pipeline
The current Autonomous Control pipeline is designed for the Trackdrive event, such that it can navigate and map an unknown race track alongside the Perceptio...
Posts by Year
2021
Tune for Disturbance Rejection for Better Performance
Update on cost tuning Over the past weeks, many tests have been run on the simulation to ensure best performance of the Husky, the final cost gains that ach...
Working MPCC on Husky Gazebo Simulation
Update Woohoo! Finally managed to get Husky Simulation to work and finish fast laps in Formula Student Germany 2018 track!
Modelling the Husky
Husky The Husky is a 4 wheeled skid-steering mobile robot that comes with its own sensors, EKF-SLAM, controller, plus various packages to support the im...
Simple Cone Positioning System
We developed a temporary Perception System that will publish the necessary Cone messages to test our Slow Lap Path Planner. This enables us to test our algor...
Stereo vision 2021 introduction
Recap of the 2020 team’s work procured two cameras and two lenses 2.3MP a2A1920-160ucBAS (Sony IMX392) colour cameras from Basler Horizon...
MUR Driverless 2021 Team
Hi everyone! Similar to last year, we are a group of 6 students from the University of Melbourne, continuing the autonomous race car project. We’ve split int...
2020
Full Stack Integration
Integration Testing
EKFSLAM performance
This is a brief update on the progress acheived on the development of ekfslam for MUR’s autonomous vehicle. There are 3 major updates to the system that wil...
LiDAR Simulation, Classification and Mounting
Featuring Ouster OS1-64
Object Detection and Inference Optimisation
In this post, I would like to discuss the various improvements we have made to our object detection pipeline. This includes our newly improved dataset and u...
LiDAR Correction
In this post, the steps taken during our LiDAR experimentation is detailed.
Mapping Module Validation
After arduous work in programming and design a basic slam implementation has been completed. Much work remains to implement and validate the system so that i...
The Communication Debate
The Piksi Multi has the capabilities to transfer data through USB, controller area network (CAN) bus, or Ethernet connections. CAN bus is the standard commun...
Stereo Vision Update
As a follow up to the previous post. It was assumed that pixel locations of the object of interest will be provided for triangulation. However, for our use c...
Real-Time Cone Detection with LiDAR
In this post, the approach taken to detect traffic cones with LiDAR will be explored and discussed. This includes the steps taken to produce the current prot...
COVID-19 GPS/IMU Alternative
As concluded in my previous post, the Piksi Multi Receiver Kit was our most favourable solution to the problem of Vehicle State Estimation (VSE) and the gath...
Point Cloud Cylindrical Volume Reconstruction
In this blog post, I would like to go over some of the steps taken to perform cylindrical volume reconstruction. For a bit of context, this particular proc...
Vehicle State Estimation Beginnings
The Vehicle State Estimation (VSE) subsystem’s role is to gather data on the vehicle’s current state through odometry and pose readings, mainly based upon it...
Object Detection and Getting Started with LiDAR
In this blog post, I will be going over some updates regarding our object detection algorithm and where we are currently at. After that, I will go over the c...
Getting Started with Stereo Vision
With our current resources, we have started with using a pair of GoPro Hero 5 sessions for our bench testing needs. By using them to capture still images of ...
Driverless SLAM
Slam The purpose of slam is to simulateneously generate the map of cones and localise the car within that track. The control team will take as an input to ra...
About MUR Driverless
We are a group of students in University of Melbourne working on an autonomous race car project. The current driverless team is split up into two main sub-sy...