Autonomous Vehicles Digest - January 2017

Consortium of automotive businesses to test C-V2X

At CES (consumer electronics show) 2017 Audi, Ericsson, Qualcomm, SWARCO Traffic Systems and the University of Kaiserslautern announced that they will be carrying out Cellular Vehicle to Everything (C-V2X) trials based on 3GPP Release 14. The trial is expected to focus on Vehicle-to-Vehicle (V2V), Vehicle-to-Infrastructure (V2I) and Vehicle-to-Pedestrian (V2P) direct communication, as well as Vehicle-to-Network (V2N) wide area communications. The aim of the trial is to show the benefits of using a unified C-V2X connectivity platform in connected vehicles and to display the range, reliability and latency advantages of C-V2X. 

Nissan autonomous car to be trialled in London

Nissan announced that it would be conducting on-road demonstrations of its autonomous driving technology in London in February 2017, showcasing the technology using a modified Nissan Leaf electric car. The company also revealed that the new Qashqai and Leaf will both be released with autonomous drive technology enabling single lane autonomy on the motorway. 

Charles de Gaulle bridge in Paris has driverless shuttle bus

From January to April 2017 people in Paris can take an autonomous shuttlebus across the Charles de Gaulle bridge, which links two stations in central Paris. The free bus, watched over by an onboard guard in case of problems, is an EZ10 made by EasyMile and will be making the crossing from 2pm to 8pm, seven days a week. The shuttle, which uses lidar sensors to assess its environment, moves across the bridge in a dedicated lane at a speed of 15mph. 

Amazon files patent for roadway management system

A patent filed by Amazon shows its solution to the problem of navigating reversible lanes (lanes that change direction dependant on traffic volume – a system commonly used in America to control traffic flow). The patent outlines a system for centralised roadway management. Information would be sent from the management system to autonomous cars about the optimum lane for highway entry, together with real-time information on the lanes’ direction status, slotting the car onto the road based on calculations of speed, occupancy, priority and other factors. Amazon states this system also has potential applications for airborne and seaborne autonomous vehicle management.

SAM – Nissan’s problem shooting system

At the Consumer Electronics Show (CES) 2017, Nissan demonstrated its solution to autonomous vehicles facing unexpected circumstances such as roadworks or accidents – SAM (Seamless Autonomous Mobility system).  Nissan says, when SAM – which is itself based on a NASA space robot data control system – meets an obstacle it will stop, radio into a command centre where a human ‘mobility manager’ will assess the situation and give advice and a safe path round the obstacle based on vehicle images and sensor data streamed from the car. The mobility manager will ‘trace’ a route for the car to follow. Once it has passed the obstacle, the car will revert to autonomous mode. The information and plan of action which has been relayed to SAM will then be transmitted to other vehicles to follow.

Mobileye, Here, BMW and Intel get it together

Mobileye, Here, BMW and Intel have joined forces to develop crowd-sourced maps. According to Here, these maps will be continuously updated. Mobileye would use its camera-based advanced driver assistance systems which would be installed on autonomous cars to collect and transmit relevant road-related data. The data would then be analysed in real-time in situ and compressed for transmission to Intel’s cloud via Intel’s planned 5G modem. Here will then use the data to update its navigation maps that BMW will use in its autonomous vehicles.

Communication is key, autonomous cars talking to human driven cars 

Researchers at the École Polytechnique Fédérale de Lausanne (EPFL) in Switzerland have developed an algorithm they claim enables autonomous cars to communicate with each other and with non-autonomous cars to avoid collisions. They equipped three vehicles – two autonomous, one manually driven – with GPS, video cameras and laser sensors and in the human-driven car, a human-machine interface to inform the driver what to do and when. In the demonstration of the algorithm at a racetrack in Sweden, the vehicles were shown to be able to communicate, their intention to merge into another vehicle’s lane and the other vehicle reacting appropriately. 

British Government sets out AV guidelines

The British Government has recently released a paper on autonomous car regulation. In the paper three main areas are covered: regulatory reform; insurance; and ADAS (Advanced Driver Assistance Systems). 
•    Regulatory reform: The Government proposes a rolling programme of reform, taking a step by step approach.
•    Insurance: The Government does not feel that any changes to product liability would be proportionate at this stage. Instead it proposes an amendment to the Road Traffic Act 1988 to include automated vehicles within its remit. It proposes a single insurer model in which the victim has a direct right against the motor insurer and the insurer has a direct right against the responsible party ie the manufacturer/software provider. This is apparently endorsed by the insurance industry. 
•    ADAS: The ban on driver’s screens displaying information unrelated to driving will remain in force until there are fully autonomous vehicles. The government intends to carry out a series of platooning trials soon to assess necessity of changes to separation distances between vehicles. 
•    Data sharing: Data from the vehicle would need to be shared to support the insurance regulation outlined above. However, it would still likely be classified as ‘personal data’ and as such legislation such as the Data Protection Act would require amending to enable data sharing with insurers.

Google’s rideshare patent

Google has recently filed a patent for determining pick-up and drop-off locations for autonomous vehicle rideshare service. The technology would allow customers to input destinations on their phones, and it would then provide the user with a predetermined pick-up destination where an autonomous car is allowed to stop. Through Google Maps the application would show users locations nearby where it would be able to pick-up or drop-off customers.

Baidu partners with BAIC Motor Corp

Chinese search engine giant Baidu has teamed up with BAIC Motor Corp, a Chinese state-owned vehicle manufacturer, to produce and promote autonomous vehicle technology. The companies announced their partnership at the Consumer Electronics Show 2017 with the stated intention of producing and trialling a level 3 autonomous vehicle by late 2017. A level 3 autonomous vehicle is one in which under certain traffic or environmental conditions drivers can shift some functions to the car to carry out autonomously. Baidu recently partnered with NVidia to combine NVidia’s self-driving computing platform with Baidu’s cloud and mapping tech. Baidu and BAIC also stated their plan to produce a BAIC made autonomous car incorporating Baidu’s telematic solutions by April 2017. 

 

 

Add this: