Raghav Mishra

Raghav Mishra

Robotics / Mechatronics Engineer

Emesent

Biography

I am a Robotics Engineer at Emesent, where I am currently working on autonomous robots for inspection, surveying and mapping which operate in inaccessible, unstructured, dirty and dangerous environments. My work spans the whole stack of perception, planning and control.

My main interests lie in robotics and control. In particular, I enjoy bringing rigorous mathematical and physical tools to making robots more safe, agile and capable. This includes non-linear system theory, optimisation and numerical optimal control, differential geometry, etc.But I enjoy a wide variety of problems from all domains of robotics and beyond.

Interests
  • Robotics
  • Control theory
  • Dynamics
  • Signal processing
  • Machine Learning
Education
  • MEng in Mechatronic Engineering

    University of Queensland

  • BEng (Hons) in Mechatronic Engineering

    University of Queensland

Experience

 
 
 
 
 
Emesent
Autonomy Engineer
Oct 2021 – Present Brisbane, QLD, AU
State estimation and autonomy for autonomous mapping systems like quadrotors and legged robots
 
 
 
 
 
University of Queensland
Teaching Assistant
Jan 2020 – Nov 2021 Brisbane, QLD, AU
Teaching Control Engineering I, Control Engineering II, Signals, Systems & Control, and Robotics & Advanced Control
 
 
 
 
 
CSIRO's Data61 - Robotics and Autonomous Systems group
Masters Postgraduate Researcher
Jan 2021 – Jul 2020 Brisbane, QLD, AU
Research into novel designs for granular jamming grippers and the use of vibrational effects on granular jamming for robotics
 
 
 
 
 
Microsoft
Software Engineering Intern
Oct 2019 – Feb 2020 Bellevue, WA, USA
Implementing distributed training of BERT neural network on Azure Compute Clusters to speed up AutoML natural language processing featurizers by distributing training to GPU compute clusters
 
 
 
 
 
CSIRO's Data61 - Robotics and Autonomous Systems group
Vacation Research Scholar
Oct 2018 – Feb 2019 Brisbane, QLD, AU
Research on improving multi-camera object detection from single camera labels for pedestrian tracking using traditional geometric features with convolutional neural networks
 
 
 
 
 
Petra Data Science
Data Science Intern
Jun 2018 – Jun 2018 Brisbane, QLD, AU
Machine learning and data science for the resources sector

Publications

(2021). Vibration Improves Performance in Granular Jamming Grippers. arXiv.

Cite

Contact