our simulation software is used across the globe
use case 1
OEM
Multinational OEM looking to accelerate AV efforts within an advanced engineering team.
Type of company: OEM
# of employees: 20,000+
Revenue: $50+ billion
Location: Primarily Michigan (but worldwide)
Example Problem:
  • Safely yielding for pedestrians and bicycles at crosswalks is a big challenge for autonomous vehicles
  • The behavior of children is difficult to predict at crosswalks
  • Often pedestrians will wait for cars to pass or wave cars through even though the pedestrian has the right-of-way
Applied Solution:
  • Creating virtual scenarios for an extremely large set of situations that autonomous vehicles may encounter near crosswalks
  • Testing in simulation that new algorithms exhibit correct, safe, and comfortable behaviors with virtual crosswalks before testing with real cars and real pedestrians
  • Running every software change through Applied's continuous integration system to ensure that each new software version is as safe or safer than the previous version
use case 2
Autonomous Vehicle Company
Silicon Valley company focused on developing self driving trucks.
Type of company: Software
# of employees: 50+
Revenue: Pre-revenue
Location: Silicon Valley
Example Problem:
  • Freight trucks need a large gap in highway traffic to safely change lanes, and lane nudging may be required to encourage other vehicles to make room for the truck
  • Nuanced truck movements coupled with the limitations of AV sensors and truck handling present a major challenge for autonomous truck lane changes
Applied Solution:
  • Simulation testing of a wide range of lane change scenarios that trucks have to deal with on the highway, including emergency situations that are very dangerous to test in the real world
  • Ensuring vehicle stability and controllability during maneuvers with various truck loads
  • Testing the limits of sensor performance and wheel traction in simulated harsh conditions
use case 3
Tier 1 Supplier
Large multinational supplier focused on using reinforcement learning to develop AV algorithms.
Type of company: Automotive supplier
# of employees: 10,000+
Revenue: $10+ billion
Location: Europe
Problem:
  • A production track Level 2 or 3 system has a tremendous number of parameters that need to be tuned to achieve good system performance and ride comfort
  • Manually tuning all parameters is a massive engineering undertaking
  • Parameter tuning needs to be repeated each time a powertrain component changes
Applied Solution:
  • Tuning parameters using learning techniques in simulation, including reinforcement learning and imitation learning
  • Refining ride comfort for new powertrain configurations and vehicle platforms in simulation before physical testing
  • Connecting simulation to machine learning frameworks and other testing tools using Applied's rich APIs

We work with all types of customers including automobile manufacturers. autonomous trucking companies. construction automators. agriculture automators. tier 1 suppliers. university labs.

request demo