April 5-7, 2022|Santa Clara Convention Center| Santa Clara, CA

Welcome to the DesignCon 2022 agenda and presentation download site. Here you can view and download conference, Chiphead Theater, and other event presentations before, during, and after the event. If you're looking for a presentation from a specific session that you're unable to find here, it is likely because the presenter has not provided permission for external use or has not yet shared their presentation with us. Please check back after the event for a more complete catalog of available presentations.

Enhancing Safety & Performance in ADAS/AV with a Motion First Approach to Perception


Joel Pazhayampallil  (Founder and CEO, BlueSpace.ai)

Location: Ballroom C

Date: Thursday, April 7

Time: 9:00 am - 9:45 am

Track: Drive World - Advanced Automotive

Format: Technical Session

Theme : Autonomous, Sensing/Vision Systems

Pass Type: 2-Day Pass, All Access Pass

Vault Recording: TBD

How can we allow for additional time and greater accuracy in understanding objects’ motion for ADAS to take effective action? Existing perception methods have several shortcomings:

  • Inability to recognize unseen/out of class objects
  • Association and tracking errors through misclassification
  • High latency, as motion estimation, needs multiple observations over time
  • Poor motion accuracy due to inference from position changes

These limitations of ADAS systems cause several false positives/false negatives and often leave the motion planning system with very little reaction time and information needed to handle safety-critical situations. As a result, the usability and trust in the current ADAS systems have been low.

In this presentation, we will discuss how a motion-first approach to perception provides the time advantage that can enable advanced ADAS features. The benefits of a motion first approach are:

  • Generic object definition by consistent motion - generalizes to all objects
  • Eliminates the need for classification and hence the need for annotation and training
  • Low Latency perception - the full-motion state is measured from a single frame
  • Motion is directly measured from doppler data - provides better accuracy by avoiding inference from the position, eliminates the need for motion priors from map data, and also eliminates the need for motion priors based on (mis)classification

Precise, low latency measurement of full-motion enables ADAS/AV with an advantage to make timely safety-critical decisions. This can finally improve the usability and trust of ADAS/AV systems and provide a smooth and safe experience for the driver and the rider.