Collision Avoidance Algorithm for Autonomous Vehicles and Robots | Rutgers University Innovation Ventures

Collision Avoidance Algorithm for Autonomous Vehicles and Robots

 

 

 

 

Invention Summary:


Autonomous navigation in a dynamic environment is a challenging task, especially when the motion of the elements populating the environment are unknown in advance and must be updated dynamically. An autonomous vehicle or robot must be able to follow a clear path towards its destination by anticipating the imminent behavior of moving obstacles and computing a safe driving path.

Researchers at Rutgers have invented a safe, and reliable collision avoidance system for robots and autonomous vehicles, which can sense and avoid obstacles accurately. The system provides better results and is faster than other solutions.

The navigation system calculates a safe and collision free velocity, that can control the movement of a robot or an autonomous vehicle and helps it to avoid any obstacle as it navigates towards its destination in a dynamic environment.

The proposed collision detection system can generate reachability and safety indexes for all potentially acceptable next step motions. More specifically, it builds upon an existing velocity obstacle algorithm, to check whether a motion will lead to a collision or not.

 

 

Advantages:

  • Fast, does not require use of computationally expensive processes (such as artificial intelligence/neural networks, optimization, sampling, or other decision-based approaches) for computing collision-free paths.
  • The algorithm does not need any training process to tune parameters.
  • Testing has shown that 100% collision free conditions can be designed for autonomous vehicles or robots.

 Market Applications:

  • Mobile robots and autonomous vehicles
  • The algorithm does not need any training process to tune parameters
  • Testing has shown that 100% collision free conditions can be designed for autonomous vehicles or robots.

Intellectual Property & Development Status:

 

The technology is patent pending and is currently available for licensing.

 

Patent Information:
Contact:
Andrea Dick
Associate Director, Licensing
Rutgers, The State University of New Jersey
848-932-4018
aid8@research.rutgers.edu
Keywords:
Robotics