Autonomous Vehicles, LEAF and others...

My Nissan Leaf Forum

Help Support My Nissan Leaf Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
cwerdna said:
cwerdna said:
AutoX sends its RoboTaxis to work in Shanghai
https://www.zdnet.com/video/autox-sends-its-robotaxis-to-work-in-shanghai/
https://www.zdnet.com/article/autox-corners-shanghai-self-driving-taxi-market/
AutoX becomes China’s first to remove safety drivers from robotaxis
https://techcrunch.com/2020/12/02/autox-removes-safety-drivers/
The company also stressed the experience it learned from “millions of miles” driven in China’s densest city centers through its 100 robotaxis in the past few years. Its rivals are also aggressively accumulating mileage to train their self-driving algorithms while banking sizable investments to fund R&D and pilot tests. AutoX itself, for instance, has raised more than $160 million to date.
Chasing Cruise and Waymo, Chinese AV company AutoX plans to begin testing in San Francisco
https://techcrunch.com/2022/02/11/chasing-cruise-and-waymo-chinese-av-company-autox-plans-to-begin-testing-in-san-francisco/amp/
 
Related to this thread...

https://insideevs.com/news/568633/herbert-diess-lidar/

What are your thoughts? I also think for now vision based autonomy is no go. Tesla phantom braking situation perhaps could be eliminated if they had Lidar verifying the existence of vehicles in front. Similarly my Ring camera gets tricked to think there is movement when sun shimmers over the water it sees. Vision only based systems impacted due to heavy rain and/or mud on camera lens.

I think Lidar (or comparable non vision based tech) is critical to the success of safe autonomy on the roads. Will be interesting to see how the industry moves on this topic.
 
Autonomous food delivery pod meets fiery end under train
This baby bot died for your dinner
https://www.autoblog.com/2022/03/04/train-hits-autonomous-robot-crossing/

The robot looks like one from https://www.starship.xyz/.
 
Waymo trials fully driverless rides in San Francisco
The company is also expanding autonomous rides to downtown Phoenix.
https://www.engadget.com/waymo-fully-driverless-rides-san-francisco-183703989.html
 
An autonomous Cruise vehicle left police confused when they tried to pull it over
https://www.engadget.com/cruise-vehicle-drives-away-from-police-194657296.html - I've only had a chance to watch part of the video.

More discussion at https://teslamotorsclub.com/tmc/threads/autonomous-car-progress.99413/page-375#post-6601637.
 
IEVS:
How Good Is Porsche's Autonomous System? On Par With Tesla?
Kyle Conner checks out the Porsche Taycan's autonomy.

https://insideevs.com/news/581636/porsche-autonomy/


When it comes to autonomy, naturally most people think of Tesla. For a long time Tesla was the undisputed king when it came to self-driving cars, however that's now beginning to change. In fact, several tests are now suggesting Tesla is no longer number one when it comes to autonomous driving - rivals are clearly beginning to up their game. . . .

Kyle's mountain test route has around 50 corners and is 15 miles long. The Taycan uses radar, unlike Tesla vehicles which now favor cameras only. The car Kyle tested came with Porsche InnoDrive including Adaptive Cruise Control and Active Lane Keep Assist, a $3,610 option.

The Taycan doesn't have a capacitive steering wheel or eye tracking. However, its not limited to pre-mapped highways. The gauge cluster only shows cars in front, not to the side. The Taycan does adapt to speed limits, however it doesn't have a speed limit offset. The safety warning system works well, with seat belts being yanked and the car coming to a complete stop with hazards on.

In summary, Kyle found the Taycan's autonomous system was impressive, although not quite as good as some other setups on the market. It works very well in corners, however there can be some issues when going straight. For the full review of the system check out the above video.
 
https://www.youtube.com/watch?v=ipc1o5M0lRQ

Pony.ai becomes first autonomous driving company to receive a taxi license in China
https://electrek.co/2022/04/25/pony-ai-becomes-first-autonomous-driving-company-to-receive-a-taxi-license-in-china/

Baidu, Pony.AI win first driverless robotaxi permits in China
https://techcrunch.com/2022/04/27/baidu-pony-ai-win-first-driverless-robotaxi-permits-in-china/
 
GCC:
Michigan State University deploys full-size autonomous electric bus

https://www.greencarcongress.com/2022/04/20220429-msu.html


Michigan State University officially has deployed its full-size autonomous electric bus (earlier post) and is accepting passengers on the 2.5-mile campus route. The autonomous bus is one of the largest of its kind to be deployed on US roadways to date.

First introduced in November 2021 during a ribbon-cutting ceremony, the bus has now completed extensive on-campus testin—including more than 650 test runs of its route spanning all hours of the day—to make the official deployment possible. As part of the process to greenlight accepting passengers, validation of the bus, route and infrastructure was granted by the National Highway Traffic Safety Administration (NHTSA).

This new mode of transportation for students, faculty, staff and visitors was made possible through a collaboration with the state of Michigan, bus manufacturer Karsan and Michigan-based ADASTEC, which focuses on delivering SAE Level-4 Automated Driving Software Platforms for commercial vehicles.

The 27-foot, 22-seat Karsan Autonomous e-ATAK bus will begin its journey each weekday morning at 9 a.m. from the MSU Commuter Lot (#89) at the intersection of Farm Lane and Mt. Hope Road. The bus’s 2.5-mile route will run nonstop, roundtrip approximately every 45 minutes from the Commuter Lot Bus Stop (#4) to the MSU Auditorium. The bus runs Monday through Thursday, 9:45 a.m. to 2:15 p.m. and Fridays from 9:45 a.m. to 12:45 p.m., with the last departures from Lot 89 of the day at 1:30 p.m. and noon respectively.

Traffic lights along the route are outfitted with intelligent roadside units and communicate with the bus to enable vehicle to infrastructure (V2I) interoperability. As an added safety measure, a licensed bus driver as a safety driver and operator from ADASTEC’s Ann Arbor office will be onboard at all times, prepared to take control instantly if needed. . . .
 
GCC:
Mercedes-Benz begins sales of DRIVE PILOT Level 3 conditionally automated driving in Germany

https://www.greencarcongress.com/2022/05/20220507-drivepilot.html


Mercedes-Benz’ DRIVE PILOT system for conditionally automated driving (SAE Level 3) can be ordered in Germany from 17 May as an optional extra for the S-Class for €5,000 and for the EQS for €7,430 (Driver Assistance Package Plus: €2,430 and DRIVE PILOT: €5,000) excl. VAT. This makes Mercedes-Benz the first car manufacturer with an international valid certification for conditional automated driving to offer such a system as an option ex-works for vehicles from series production.

DRIVE PILOT enables customers to hand the driving task over to the system under certain conditions in heavy traffic or congestion situations on suitable motorway sections in Germany up to a speed of 60 km/h (37 mph).

After activating DRIVE PILOT, the system controls the speed and distance, and guides the vehicle within its lane. The route profile, events occurring on the route and traffic signs are all analysed and taken into consideration. DRIVE PILOT also reacts to unexpected traffic situations and handles them independently, e.g. by means of evasive manoeuvres within the lane or by using braking maneuvers.

Mercedes-Benz’s system for conditionally automated driving (SAE-Level 3) builds on the vehicle sensing technology of the Driving Assistance Package and includes additional sensors that the manufacturer considers indispensable for safe operation. These mainly include radar, LiDAR and cameras, but ultrasound and moisture sensors also provide valuable data.

In parallel to the successful launch of DRIVE PILOT in Germany, Mercedes-Benz aims to obtain regulatory series approval for the two US states of California and Nevada by the end of the year, provided the legal situation permits the system operation. . . .

After activating DRIVE PILOT, the system controls the speed and distance, and guides the vehicle within its lane. The route profile, events occurring on the route and traffic signs are all analysed and taken into consideration. DRIVE PILOT also reacts to unexpected traffic situations and handles them independently, e.g. by means of evasive manoeuvres within the lane or by using braking maneuvers.

Mercedes-Benz’s system for conditionally automated driving (SAE-Level 3) builds on the vehicle sensing technology of the Driving Assistance Package and includes additional sensors that the manufacturer considers indispensable for safe operation. These mainly include radar, LiDAR and cameras, but ultrasound and moisture sensors also provide valuable data.

In parallel to the successful launch of DRIVE PILOT in Germany, Mercedes-Benz aims to obtain regulatory series approval for the two US states of California and Nevada by the end of the year, provided the legal situation permits the system operation. . . .

In the unlikely event of a malfunction, the vehicle remains maneuverable due to its redundant system design, allowing DRIVE PILOT to perform a safe handover to the driver. If the driver does not comply with the takeover request within the maximum allotted time of ten seconds, for example due to a medical emergency, DRIVE PILOT promptly initiates an emergency stop that is safe both for the vehicle and the following traffic.

For this purpose, the Intelligent Drive Controller continuously calculates the optimum trajectory for coming to a safe stop. Meanwhile, the maneuverability of DRIVE PILOT is designed to keep the vehicle within its lane and avoid collisions with other road users and objects on the road. . . .

Mercedes-Benz is ensuring that the new technology is being put on the road in a legally compliant manner. Engineers, lawyers, compliance managers, data protection officers and ethics experts have all worked together in the development process.

One basis for automated driving is provided by additional control mechanisms, for example: For safety-relevant functions such as pedestrian detection, the Mercedes-Benz’ engineers deliberately do not use algorithms, for example through self-learning approaches. The focus is more on what is known as “supervised learning”—i.e., Mercedes-Benz defines and control what the artificial intelligence is allowed to learn.

Before the AI software is used on the roads, extensive validation testing is carried out to ensure that the AI works as desired in real traffic conditions. In the case of pedestrian detection, AI helps the system to identify objects and situations on and next to the road quickly and safely. The car manufacturer’s ethical requirement is that the detection process is non-discriminatory. This means that the vehicle’s various sensors permanently monitor the road and the roadside, to always detect people correctly, regardless of their clothing, body size, posture or other characteristics.

With the opening up of the Road Traffic Act (StVG) for Level 3 systems in 2017, Germany was the first country to create a legal basis for the intended use of these systems. The technical approval regulation with which such a system can be certified did not come into force until the beginning of 2021. Since then, it can be implemented in Europe. Mercedes-Benz is the first automotive company in the world to meet the demanding legal requirements of UN Regulation No. 157 for a Level 3 system.

In detail, this results in requirements for the vehicle and duties for the driver: In conditional automated driving mode, the vehicle must master the driving task safely and comply with all traffic regulations. The driver still has duties in public road traffic, in particular to comply with other traffic regulations. To this end, the driver must remain ready to take over and resume control when requested by DRIVE PILOT or due to obvious circumstances.
 
GCC:
Covenant and Aurora to explore integrating Aurora Horizon autonomous trucking for long-haul freight

https://www.greencarcongress.com/2022/05/20220512-covenant.html


Aurora Innovation Covenant Logistics Group are collaborating to explore ways t#o optimize Covenant’s long-haul operations with Aurora’s autonomous trucking product. The two companies will explore the integration and deployment of Aurora Horizon within Covenant’s operations.

Part of Covenant’s diversified services is long-haul trucking, making the benefits of autonomous technology in its network especially attractive. A hybrid model that blends trucks driven by humans with trucks driven autonomously would allow Covenant to realize the best of both and potentially expand its business with a combination of rapid, high-efficiency autonomous transport on long-haul routes and flexible human navigation at the endpoints.

The collaboration will initially focus on driving value, efficiency, and education by:

Identifying the lanes where the Aurora Driver can make the biggest impact while improving driver quality of life.

Optimizing fuel efficiency and maximizing the up-time of Covenant’s fleet with Aurora Horizon, built for 24/7/365 operation.

Collaborating on how to educate the transportation and logistics industry effectively on the benefits of autonomous technology, especially on long-haul lanes. . . .

Covenant Logistics Group, Inc., through its subsidiaries, offers a portfolio of transportation and logistics services to customers throughout the United States. Primary services include asset-based expedited and dedicated truckload capacity, as well as asset-light warehousing, transportation management, and freight brokerage capability.

In addition, Transport Enterprise Leasing is an affiliated company providing revenue equipment sales and leasing services to the trucking industry.

The Aurora Driver is a self-driving system designed to operate multiple vehicle types, from freight-hauling semi-trucks to ride-hailing passenger vehicles, and underpins Aurora Horizon and Aurora Connect, its driver-as-a-service products for trucking and ride-hailing. Aurora is partnered with industry leaders across the transportation ecosystem, including Toyota, FedEx, Volvo Trucks, PACCAR, Uber, Uber Freight, US Xpress, and Werner.


Given driver shortages and the lack of need for transshipment outside of loading docks for much of the long-haul trucking business, plus the majority of the trips are on/adjacent to freeways, long-haul trucks seem like a fairly early use case for autonomy. They'll have to figure out things like providing chain installation and removal and suchlike, but freeways have the least traffic conflicts so should be the easiest to handle.
 
San Francisco Police Are Using Driverless Cars as Mobile Surveillance Cameras
“Autonomous vehicles are recording their surroundings continuously and have the potential to help with investigative leads,” an internal training document states.
https://www.vice.com/en/article/v7dw8x/san-francisco-police-are-using-driverless-cars-as-mobile-surveillance-cameras
 
Ford-backed robotaxi start-up Argo AI is ditching its human safety drivers in Miami and Austin
https://www.cnbc.com/2022/05/17/argo-ai-robotaxis-ditch-human-safety-drivers-in-miami-and-austin.html
"Robotaxi start-up Argo AI said Tuesday it has begun operating its autonomous test vehicles without human safety drivers in Miami and Austin, Texas.
For now, the driverless vehicles will shuttle Argo AI employees, not paying passengers."
 
Pony.ai’s license to test autonomous vehicles with safety drivers in California is revoked
The DMV found ‘numerous safety violations’ on the driving records of the company’s safety drivers
https://www.theverge.com/2022/5/26/23141775/pony-ai-revoke-license-california-av-testing-dmv
 
Cruise is now charging for rides in its driverless vehicles in San Francisco
https://www.theverge.com/2022/6/23/23180156/cruise-driverless-vehicle-charge-riders-san-francisco
 
Tesla is cutting about 200 Autopilot jobs and closing office in San Mateo, California
https://www.cnbc.com/2022/06/28/tesla-cutting-200-jobs-closing-autopilot-office-in-san-mateo.html
 
Self-driving cars from GM’s Cruise block San Francisco intersection in latest problem for autonomous vehicles
https://www.cnbc.com/2022/07/01/self-driving-cars-from-gms-cruise-block-san-francisco-streets.html
 
Autonomous vehicles startup Nuro winds down operations in Phoenix
https://www.yahoo.com/video/autonomous-vehicles-startup-nuro-winds-190237796.html

Ford-backed autonomous car startup Argo AI lays off 150 employees
The layoffs make up 6 percent of Argo AI’s team
https://www.theverge.com/2022/7/9/23201966/ford-backed-autonomous-car-startup-argo-ai-lays-off-150-employees-driverless

Whistleblower Says GM's Cruise Isn't Taking Safety Seriously Enough
Anonymous employee claims team didn't feel service was launch-ready, "but there is fear of admitting this because of expectations from leadership, investors."
https://jalopnik.com/whistleblower-says-gms-cruise-isnt-taking-safety-seriou-1849182155

This burning RV achieved FSD (Fiery Self Driving). Skip to 3:25. I haven't watched the whole thing, only about the first half.
https://www.theautopian.com/watching-a-burning-rv-drive-itself-into-trees-is-alarming-but-theres-an-explanation-that-doesnt-involve-ghosts/
 
Behind the scenes of Waymo’s worst automated truck crash
Self-driving semi was hit and run in May, then everything went quiet
https://techcrunch.com/2022/07/01/behind-the-scenes-of-waymos-worst-automated-truck-crash

TuSimple reportedly tried to pass off a self-driving truck crash as ‘human error
The autonomous truck startup is being investigated for the April incident
https://www.theverge.com/2022/8/4/23288794/tusimple-self-driving-truck-crash-investigation
 
Back
Top