Hardware Podcasts

Who’s Actually Driving This Thing? | Really, Really Long Trunks

EE Times Current · 12 Feb 2021 · 29m

What this episode covers

Missy Cummings, a professor at Duke University and former Navy fighter pilot, explains the concept of 'mode confusion' in autonomous vehicles, where drivers mistakenly believe the car is in control when it is not. She critiques Tesla's Autopilot for inconsistent alerting and lack of clear state transitions, citing her own testing that showed extreme variability in how the system initiates warnings. Cummings argues that automakers must implement standardized, multi-modal alerts (visual, audio, and haptic) to clearly signal when automation disengages, drawing parallels to aviation safety protocols. She also dismisses current torque-based driver monitoring systems as ineffective proxies for attention, advocating instead for camera-based systems and last-second threat detection like AEB.

Most relevant to

early-stage companies

Topics

Mentioned

Companies
Tesla, Cadillac, Airbus, Waymo
Also
Duke University

Related episodes from other shows

Other shows on the same ground, closest first.

  • The Robot Report Podcast · 1h 14m

    Passenger of confused Waymo robotaxi discusses what went wrong

    Joel Johnson, a prolific Waymo One rider and content creator, discusses a specific incident where his robotaxi became confused by construction cones in Chandler, Arizona, leading to multiple disengagements and miscommunication with remote assistance. Tom Ryden, executive director of MassRobotics, explains the development of a new mobile robot interoperability standard designed to allow autonomous mobile robots (AMRs) from different vendors to coordinate and share data in shared environments like warehouses.

    Robotics & autonomyDebugging & failure analysis
    7 chapters

  • The Next Byte Podcast · 23m

    134. Can We Trust Self-Driving Cars? 🚗💥

    Hosts Daniel and Forbode discuss a proposal by MIT researchers to bridge the gap between autonomous vehicle technologists and regulators by introducing a hybrid safety model. They analyze how remote human supervisors can assist self-driving cars with complex tasks like highway merging, achieving 99.999% reliability when 30% of vehicles are cooperative. The episode highlights the scalability challenges of this approach, noting that while feasible for early adoption, it requires significant labor resources compared to fully autonomous systems.

    Robotics & autonomyValidation & product-market fit
    7 chapters

  • How I Built This · 41m

    The future of driving is autonomous with Dmitri Dolgov of Waymo

    Dmitri Dolgov, co-CEO of Waymo, recounts the company's decade-long journey from a secret Google project to the world's first fully autonomous ride-hailing service. He details the technical evolution from the early DARPA Challenge vehicles to the current Jaguar I-PACE fleet, explaining why Waymo chose a multi-sensor approach with LiDAR and radar over Tesla's camera-only vision. Dolgov also discusses the strategic pivot to full autonomy, the safety framework that underpins their operations, and the business model involving Waymo One, Uber partnerships, and future personal vehicle ownership.

    Robotics & autonomySensors & measurement in products
    6 chapters

  • How I Built This · 40m

    When trucks drive themselves with Chris Urmson of Aurora

    Chris Urmson, founder of Aurora, discusses his journey from the DARPA Grand Challenges to leading Google's self-driving car project and finally launching Aurora to automate long-haul trucking. He explains Aurora's 'First Light' LIDAR technology, the distinction between Level 2 assistance and Level 4 autonomy, and the company's strategy of selling an autonomous driving system rather than building vehicles. Urmson outlines the safety case, regulatory engagement, and the phased rollout plan for driverless trucks on specific US freight corridors.

    Starting a product companyMicrocontrollers & dev platforms
    6 chapters

  • The Next Byte Podcast · 20m

    123. Augmented Bad Drivers Turbocharge Autonomous Vehicle Development

    Hosts Daniel and Forbode discuss research from the University of Michigan that uses mixed reality to accelerate autonomous vehicle (AV) testing. By superimposing virtual 'bad driver' vehicles onto real physical cars on the MCity test track, researchers can simulate rare safety-critical events without real-world risk. This approach, leveraging data from smart intersections in Ann Arbor and Detroit, reduces the required testing miles by 99.99%, addressing the 'curse of rarity' in AV training data. The hosts highlight how this method lowers costs and time while improving safety for companies like Waymo, Tesla, and Cruise.

    Connected products & IoTPrototyping & iteration
    9 chapters

The summary, chapters and topics are generated from the episode’s audio by this directory, not written by EE Times Current. The episode itself belongs to its publisher.