Self-Driving Cars with Fexingo: Autonomous Vehicles, Lidar, and Mobility Tech

Episode 29 of Self-Driving Cars with Fexingo: Lucas and Luna dig into one of the most stubborn unsolved problems in autonomous driving — the four-way stop intersection. While robotaxis handle highways and simple traffic lights with ease, multi-way stops force human-like negotiation, eye contact, and social rules that AI still struggles with. Lucas explains why Waymo and Cruise vehicles sometimes f...

Episode 28 of Self-Driving Cars with Fexingo looks at a surprisingly human problem in autonomous mobility: lost items. When Uber published data showing thousands of items left behind in robotaxis — from Squishmallows to dentures — it revealed how the driverless experience changes the recovery process. Lucas and Luna discuss why robotaxis lose more items than human-driven cars, how companies are bu...

Episode 27 of Self-Driving Cars with Fexingo dives into the persistent challenge of construction zones for autonomous vehicles. Lucas and Luna discuss why even the most advanced robotaxis and AVs still struggle to navigate temporary road layouts, cones, flaggers, and conflicting signage. They anchor the conversation with a recent incident involving a Waymo vehicle in San Francisco and examine the...

Episode 26 explores the hidden workforce behind autonomous vehicles: remote operators who take control when the AI gets confused. Lucas and Luna break down the economics of this emerging job market, using data from Rivian's 17.8% stock surge and Ford's 8.6% gain to illustrate investor interest in human-in-the-loop systems. They discuss how companies like Waymo and Cruise are scaling operations cen...

In this episode of Self-Driving Cars with Fexingo, Lucas and Luna explore a paradigm shift in autonomous navigation: the move away from high-definition pre-mapped roads toward vehicles that learn to drive in unfamiliar environments in real time. They focus on a specific breakthrough from a team at MIT CSAIL, published in early 2026, where a single vehicle navigated a 10-kilometer stretch of unmark...

Lucas and Luna explore why autonomous vehicle companies are quietly abandoning years of high-definition map data and starting over. Real-time mapping startups like DeepMap and Wayve are challenging legacy approaches from HERE and TomTom. The hosts discuss the shift from centimeter-precise pre-mapped roads to self-healing maps that update themselves, and why lidar-on-a-stick mapping vans may soon b...

In this episode of Self-Driving Cars with Fexingo, Lucas and Luna explore why traditional road-testing approaches are failing for autonomous vehicles. They discuss how simulation-first validation is becoming the industry standard, referencing Waymo's 25 billion simulated miles, and examine the shift from miles-per-disengagement to scenario-based safety metrics. The hosts also touch on how this cha...

In this episode, Lucas and Luna explore why autonomous vehicle companies are increasingly designing their own custom chips instead of relying on off-the-shelf components. Using NVIDIA's dominance as a backdrop, they examine how Tesla, Waymo, and Chinese startups are investing in bespoke silicon to gain performance and cost advantages. The conversation touches on the trade-offs between flexibility...

Episode 21 of the Self-Driving Cars with Fexingo podcast digs into the unglamorous backbone of autonomous vehicle fleets: remote human operators. Hosts Lucas and Luna explore why every major robotaxi service, from Waymo to Cruise to Zoox, still has a human-in-the-loop watching from a desk hundreds of miles away. They break down the real numbers: how many vehicles each operator can monitor, what ha...

Episode 20 of Self-Driving Cars with Fexingo dives into the insurance data crisis facing autonomous vehicle fleets. Lucas and Luna explore why traditional actuarial models break down when machines drive, how a single accident in San Francisco reshaped policy, and the surprising role of simulation data from companies like Waymo and Cruise. With context from recent market moves — Rivian up 14.6 perc...

Episode 19 of Self-Driving Cars with Fexingo tackles the overlooked bottleneck in autonomous vehicle deployment: the humans in remote monitoring centers. Lucas and Luna explore how the promise of full autonomy has given way to a reality where fleets require one remote operator for every few robotaxis, creating a new labor-intensive industry. They discuss scaling challenges, the economics of operat...

Every major self-driving company has promised autonomous trucks on U.S. highways by 2026. So why do most trucks still have a safety driver behind the wheel? Lucas and Luna dig into the hard operational problem that regulators and engineers are only now admitting: the last mile. Waymo's autonomous vehicle registrations outpace Tesla's 4 to 1 this year, yet Waymo has pulled all its trucks off public...

Episode 17 of Self-Driving Cars with Fexingo explores a hidden bottleneck slowing autonomous vehicle development: data center energy consumption. Lucas and Luna break down how training a single full-stack autonomous driving model can consume as much electricity as 100 American homes use in a year. They discuss the struggle of companies like Waymo and Cruise to secure enough compute power, the shif...

Episode 16 of Self-Driving Cars with Fexingo dives into the one robotaxi fleet that has turned a profit in 2026. We look at Baidu's Apollo Go in Wuhan, which hit operational breakeven in Q1 2026 by running 4,000 driverless cars in a dense, predictable city. Lucas and Luna break down the unit economics: $0.40 per mile operating cost versus $0.65 per mile in revenue, with utilization above 60%. They...

This episode of Self-Driving Cars with Fexingo examines a surprising trend in May 2026: traditional automakers like GM and Ford are surging ahead of Tesla in autonomous vehicle technology. Lucas and Luna break down the market data showing Ford up 20% and GM up 10% in the past five days, while Tesla lags. They explore how legacy automakers are leveraging partnerships with lidar startups and cautiou...

Episode 14 of Self-Driving Cars with Fexingo looks at an unexpected consequence of the autonomous vehicle boom: a growing market for used self-driving cars. Lucas and Luna dig into the mechanics of how robotaxi fleets retire vehicles after 300,000 miles in three years, the residual value problem facing early-generation autonomous cars, and the surprising secondary buyers — from university research...

Lucas and Luna explore the surprisingly stubborn challenge of finding talent for self-driving car companies. They break down why autonomous vehicle startups have quietly become some of the most desperate hirers in tech, how a single lidar calibration engineer can command a $400,000 salary, and the structural mismatch between the skills available and the skills needed. They also discuss what this m...

Lucas and Luna trace the astonishing cost reduction in lidar sensors — from a $75,000 unit on the early DARPA Grand Challenge vehicles to today's $500 solid-state sensors. They unpack the manufacturing innovations, the role of Chinese supply chains, and what this means for the self-driving industry's timeline. Along the way, they reference Rivian's 6.5% weekly gain and Ford's 14.6% surge as signal...

In this episode of Self-Driving Cars with Fexingo, Lucas and Luna explore why the autonomous vehicle industry's shift to purpose-built silicon is creating a hidden fragmentation crisis. They break down how companies like NVIDIA, Qualcomm, and Tesla are each betting on different chip architectures, making it harder for software to scale across platforms. With references to recent market moves and t...

Self-driving cars have mastered sunny Phoenix but still struggle with rain, fog, and snow. In this episode, Lucas and Luna look at how robotaxi operators — from Waymo to Chinese operators like Baidu — are tackling adverse weather in 2026. They discuss sensor fusion strategies, the role of thermal cameras, and why one startup is training neural networks on synthetic raindrops. Plus, they check the...

Autonomous vehicles are hitting real roads in 2026, but there's a massive regulatory gap threatening deployment. Lucas and Luna dig into how the federal government has essentially punted oversight to states, creating a chaotic patchwork that companies like Waymo and Cruise have to navigate. They break down the specific split—the NHTSA handles vehicle design, but states regulate operations—and why...

Episode 8 zooms into the invisible bottleneck for autonomous vehicle development: how companies build, label, and license the high-definition ground-truth data sets used to train their perception systems. Lucas and Luna explore why data is becoming a proprietary moat, how lidar-camera fusion changed labeling workflows, and the surprising rise of synthetic data from game-engine simulations. With Ri...

Everyone talks about lidar and software when it comes to self-driving cars, but the real bottleneck might be the chips inside. Lucas and Luna dig into why autonomous-vehicle computing demands are exploding and how a few semiconductor players are quietly becoming the gatekeepers of the industry. They look at what an autonomous vehicle actually needs to process per second, why legacy auto suppliers...

Lucas and Luna tackle the messy insurance liability gap that autonomous vehicle companies face in 2026. With robotaxis and autonomous trucks operating in limited geographies, traditional auto insurance models break down. Lucas explains why Tesla's recent pivot to lidar doesn't fix the liability question, and digs into how a Berlin startup called Peec is quietly building the data infrastructure for...

Episode 5 of Self-Driving Cars with Fexingo dives into the software bottleneck that's delaying autonomous trucking deployments. Lucas and Luna examine why the sensor stack has largely been solved but the planning and decision-making software still struggles with edge cases like highway merges and construction zones. They reference a recent McKinsey report showing autonomous trucking could add $1.5...