Things to Do in L.A. Tap to enable a layout that focuses on the article. 'Waymo problems': Man jumps into trunk of driverless taxi in L.A., gets stuck and is removed by police This is read by an automated voice. Please report any issues or inconsistencies here . A man hopped into the open trunk of a Waymo in L.A. only to get stuck inside. Police removed the man after the next Waymo passenger discovered him in the trunk.

Decentralized (PO)MDPs provide an expressive framework for sequential decision making in a multiagent system. Given their computational complexity, recent research has focused on tractable yet practical subclasses of Dec-POMDPs. We address such a subclass called C Dec-POMDP where the collective behavior of a population of agents affects the joint-reward and environment dynamics. Our main contribution is an actor-critic (AC) reinforcement learning method for optimizing C Dec-POMDP policies. V anilla AC has slow convergence for larger problems. To address this, we show how a particular decomposition of the approximate action-value function over agents leads to effective updates, and also derive a new way to train the critic based on local reward signals. Comparisons on a synthetic benchmark and a real world taxi fleet optimization problem show that our new AC approach provides better quality solutions than previous best approaches.

Neural Information Processing Systems http://nips.cc/

In this paper we introduce a novel deictic object-oriented framework that has provably efficient learning bounds and can solve a broader range of tasks.

Online ride-hailing platforms aim to deliver efficient mobility-on-demand services, often facing challenges in balancing dynamic and spatially heterogeneous supply and demand. Existing methods typically fall into two categories: reinforcement learning (RL) approaches, which suffer from data inefficiency, oversimplified modeling of real-world dynamics, and difficulty enforcing operational constraints; or decomposed online optimization methods, which rely on manually designed high-level objectives that lack awareness of low-level routing dynamics. To address this issue, we propose a novel hybrid framework that integrates large language model (LLM) with mathematical optimization in a dynamic hierarchical system: (1) it is training-free, removing the need for large-scale interaction data as in RL, and (2) it leverages LLM to bridge cognitive limitations caused by problem decomposition by adaptively generating high-level objectives. Within this framework, LLM serves as a meta-optimizer, producing semantic heuristics that guide a low-level optimizer responsible for constraint enforcement and real-time decision execution. These heuristics are refined through a closed-loop evolutionary process, driven by harmony search, which iteratively adapts the LLM prompts based on feasibility and performance feedback from the optimization layer. Extensive experiments based on scenarios derived from both the New York and Chicago taxi datasets demonstrate the effectiveness of our approach, achieving an average improvement of 16% compared to state-of-the-art baselines.

Driverless taxis from Waymo will be on London's roads next year, US firm announces Wed 15 Oct 2025 05.00 EDTLast modified on Wed 15 Oct 2025 05.02 EDT Driverless taxis from Waymo will be available for hire on London's roads next year, the US company has announced. The UK capital will become the first European city to have an autonomous taxi service of the kind now familiar in San Francisco and four other US cities using Waymo's technology. Waymo said its cars were now on their way to London and would start driving on the capital's streets in the coming weeks with "trained human specialists", or safety drivers, behind the wheel. The company - originally formed as a spin-off from Google's self-driving car programme and part of the same parent group, Alphabet - said it would scale up operations and work closely with the Department for Transport and Transport for London to obtain the necessary permissions to offer fully autonomous rides in 2026. Uber and the UK tech company Wayve have also announced their own plans to trial their driverless taxis in the capital next year, after the British government said it would accelerate rules allowing public trials to take place before legislation enabling self-driving vehicles passes in full.

Self-driving Ubers are expected to appear on roads in London next year after the government said trials of fully autonomous vehicles would be brought forward to spring 2026. Companies will be allowed to run pilots of small-scale taxi or "bus-like" services for public use – and, for the first time in Europe, without any human safety driver onboard or in the driving seat. Uber will partner with the UK tech firm Wayve to launch trials of taxis bookable via its app in the capital, its largest European market. A fuller rollout of self-driving taxis, or robotaxis, will come after the Automated Vehicles Act fully takes effect in late 2027. The UK has sped up the process now that driverless taxis have become established in San Francisco in the US and numerous cities in China. Uber rolled out its first driverless taxis with the US firm Waymo in Austin, Texas, in March this year, where Tesla is also planning to launch a rival autonomous service this month.

As Los Angeles police struggled with another day of unrest in downtown L.A., several Waymo autonomous taxis were set on fire, sending black smoke billowing into the air. The dramatic images were captured during an afternoon of clashes between large groups who were protesting immigration raids by the Trump administration and L.A. police who were trying to maintain order. For some time, protesters blocked traffic on the 101 Freeway before California Highway Patrol officers slowly pushed them back. Police advised residents to avoid the the 101 Freeway through downtown L.A. Images of the Waymo cars on fire on Los Angeles Street were broadcast nationally as Los Angeles has become a flashpoint in the immigration debate. Tires were slashed, windows smashed, and anti-ICE messages spray-painted over the cars, which were parked in a row.