Chapter
Solving Autonomous Vehicles through Data-Driven Approaches and Model-Based Approaches
32:04 - 38:43 (06:39)
The use of data-driven and learning-based approaches in combination with model-based systems is essential to successfully develop autonomous vehicles. However, identifying all corner cases using a single sensing modality and machine learning alone is a tough challenge.
Clips
The greater challenge for autonomous vehicles is to identify all the corner cases using a single sensing modality and machine learning alone.
32:04 - 36:10 (04:06)
Summary
The greater challenge for autonomous vehicles is to identify all the corner cases using a single sensing modality and machine learning alone. Therefore, data-driven approaches must be utilized with traditional model-based solutions to create a more comprehensive and reliable perception system.
ChapterSolving Autonomous Vehicles through Data-Driven Approaches and Model-Based Approaches
EpisodeVijay Kumar: Flying Robots
PodcastLex Fridman Podcast
The relationship between the amount of data required for high accuracy in machine learning and the energy consumption for running data farms or servers is an exponential curve.
36:10 - 37:15 (01:05)
Summary
The relationship between the amount of data required for high accuracy in machine learning and the energy consumption for running data farms or servers is an exponential curve.
ChapterSolving Autonomous Vehicles through Data-Driven Approaches and Model-Based Approaches
EpisodeVijay Kumar: Flying Robots
PodcastLex Fridman Podcast
The speaker discusses the challenges of implementing autonomous technology in both driving and flight, but notes that flight has several advantages due to pre-programmed trajectories that are less likely to encounter obstacles.
37:15 - 38:43 (01:27)
Summary
The speaker discusses the challenges of implementing autonomous technology in both driving and flight, but notes that flight has several advantages due to pre-programmed trajectories that are less likely to encounter obstacles.