The Data Science behind Cruise’s Robotaxi Fleet
summary
Describes concrete systems Cruise uses: dense safety metrics derived from near collisions and human accident benchmarks, simulation test case sampling for release decisions, and real fleet telemetry streaming to the cloud while cars retain local navigation. Commercial work includes city zoning with time discretization and a linear program for transition rates to position vehicles, a discretized charge-level LP for charging and deployment schedules, pickup spot optimization trading walk time, trip time and stop failure probability, and road-specific risk scores used in routing.
AI-written summary · about
chapters (8)
- 0:00Cruise’s robotaxi data science
- 1:04Two data science areas
- 3:39Commercial challenges
- 4:49Vehicle positioning
- 7:58Charging and maintenance
- 9:45Pickup and drop-off optimization
- 11:27Risk-based routing
- 12:55Q&A
related talks
34mData Drives Business
28mPredicting the Unpredictable Cruise’s Continuous Learning Machine
08mdata science in automotive
21mML Infrastructure for Autonomous Vehicles
23mOptimising organized transport at scale
26mData Driven Culture in Software Defined Vehicles
23mMaking Sense of Streaming Sensor Data: How Uber Detects On-trip Car Crashes
38mSampling and Modelling the World in Real-Time for Mobility