The paper presents a semi-parameteric model for long-term planning in a general space of problems. It works by training parametric goal-conditioned policies accurate only on local distances (i.e. when current state and goal are within some distance threshold) and leveraging the replay buffer to non-parametrically sample a graph of landmarks which the local goal-conditioned policy can accurately produce paths between. Moving to any goal state is then accomplished by (1) moving to the closest landmark using the goal-conditioned policy, (2) planning a path to the landmark closest to the goal using value-iteration on the (low-dimensional) graph of landmarks, (3) using the goal-conditioned policy to get to the goal state from the closest landmark. The paper essentially tackles the problem that goal-conditioned policies, or Universal Value Function Approximators (UVFA), degrade substantially in performance as the planning horizon increases. By leveraging the replay buffer to provide way-points for the algorithm to plan locally along, accuracy over longer ranges is maintained.

When a customer said they had driven across three US states, from Texas to North Carolina, to get their car fixed, Tesla servicer Jason Hughes knew something must be up. It turned out to be an unusual problem: the Model S had lost a third of its battery range in an instant, while it was parked on a driveway. One of the US electric carmaker's big selling points is that it is constantly connected to its vehicles via mobile networks, offering software updates and entertainment downloads "over the air", or OTA. That remote connection could be revolutionary for the industry, opening the door for self-driving features to be downloaded and TV shows to be livestreamed in high definition. But the connected car is having teething problems.

When it comes to robocars, new LIDAR products were the story of CES 2018. Far more companies showed off LIDAR products than can succeed, with a surprising variety of approaches. CES is now the 5th largest car show, with almost the entire north hall devoted to cars. In coming articles I will look at other sensors, software teams and non-car aspects of CES, but let's begin with the LIDARs. When it comes to robocar LIDAR, the pioneer was certainly Velodyne, who largely owned the market for close to a decade. Their $75,000 64-laser spinning drum has been the core product for many years, while most newer cars feature their 16 and 32 laser "puck" shaped unit.

BMW is an iconic brand in the automobile industry, well-established and well-liked, but that hasn't translated to success for the German automaker when it comes to electric vehicles. Having sold only about 25,000 of the i3 hatchbacks in 2015, the company is planning to upgrade its small car next year. German weekly Welt am Sonntag (in German) reported Sunday the i3 will get a makeover that will see both the front and rear ends of the car reworked. Additionally, new battery technology will give the vehicle increased range as well, which BMW says is currently 114 miles on full charge. The range can be extended up to 180 miles using a "Range Extender."