Claudio Ruch

Last Name

Ruch

First Name

Claudio

ORCID

0000-0002-0511-474X

Show all metadata

Search Results

Publications 1 - 10 of 14

The +1 Method: Model-Free Adaptive Repositioning Policies for Robotic Multi-Agent Systems
Ruch, Claudio; Gächter, Joel; Hakenberg, Jan; et al. (2020)
IEEE Transactions on Network Science and Engineering
Robotic multi-agent systems can efficiently handle spatially distributed tasks in dynamic environments. Problem instances of particular interest, and generality are the dynamic traveling repairman problem, and the dynamic vehicle routing problem. Operational policies for robotic fleets solving these two problems take decisions in an online setting with continuously arriving demands to optimize service level, and efficiency, and can be classified along several lines. First, some require a model of the demand, e.g., based on historical information, while others work model-free. Second, they are designed for different operating conditions from light to heavy system load. Third, they work in a time-invariant or time-varying setting. We present a novel class of model-free operational policies for time-varying demands, which have performance independent of the load factor, a combination of properties not achieved by other operational policies in the literature. The underlying principle of the introduced policies is to send available robots to recent service request locations. In simple terms, they rely on sending more than one robot for every service request arriving to the system. This leads to an advantage in scenarios where demand is non-uniformly distributed, and correlated in space, and time. We provide performance guarantees for both the time-invariant, and the time-varying cases as well as for correlated demand. We verify our theoretical results numerically. Finally, we apply our operational policy to the problem of mobility-on-demand fleet operation, and demonstrate that it outperforms model-based, and complex algorithms across all load ranges, despite its simplicity. © 2020 IEEE.
Improved public transportation in rural areas with self-driving cars
Sieber, L.; Ruch, Claudio; Hörl, Sebastian; et al. (2020)
Transportation Research Part A: Policy and Practice
Improved public transportation in rural areas with self-driving cars: The Example of Swiss Train Lines
Sieber, Lukas; Ruch, Claudio; Hörl, Sebastian; et al. (2019)
Arbeitsberichte Verkehrs- und Raumplanung
Public transport lines, especially train lines, have historically played an important role as economic lifelines of rural areas. They are one of the most important factors contributing to economic prosperity as they provide access to mobility for all the inhabitants of these regions. Maintaining such rural public transport lines can be a challenge due to the low utilization inherent to rural areas. Today, with the emergence of fully self-driving cars, on-demand mobility schemes in which autonomous robotic taxis transport passengers, are becoming possible. In this work, we analyze if rural public transport lines with low utilization can be replaced with autonomous mobility-on-demand systems. More specifically, we compare the existing public transportation infrastructure to a hypothetical autonomous mobility-on-demand system both in terms of cost and service level. We perform our analysis using an agent based simulation approach in which unit capacity robotic taxis are operated in a street network taking into account congestion effects and state-of-the-art control (dispatching and rebalancing) strategies. Our study targets the case of four rural train lines in Switzerland that operate at low utilization and cost coverage. We show that a unit-capacity mobility-on-demand service with self-driving cars reduces both travel times and operational cost in three out of four cases. In one of the three cases, even a service with human driven vehicles would provide higher service levels at lower cost. The results suggest that centrally coordinated mobility-on- demand schemes could be a very attractive option for rural areas.
Integrating Autonomous Mobility-on-Demand Systems in Existing Cities
Ormezzano, Nicolo; Ruch, Claudio; Frazzoli, Emilio (2019)
Scalable Model Predictive Control for Autonomous Mobility-on-Demand Systems
Carron, Andrea; Seccamonte, Francesco; Ruch, Claudio; et al. (2021)
IEEE Transactions on Control Systems Technology
Technological advances in self-driving vehicles will soon enable the implementation of large-scale mobility-on-demand (MoD) systems. The efficient management of fleets of vehicles remains a key challenge, in particular to achieve a demand-aligned distribution of available vehicles, commonly referred to as rebalancing. In this article, we present a discrete-time model of an autonomous MoD system, in which unit capacity self-driving vehicles serve transportation requests consisting of a (time, origin, destination) tuple on a directed graph. Time delays in the discrete-time model are approximated as first-order lag elements yielding a sparse model suitable for model predictive control (MPC). The well-posedness of the model is demonstrated, and a characterization of its equilibrium points is given. Furthermore, we show the stabilizability of the model and propose an MPC scheme that, due to the sparsity of the model, can be applied even to large-scale cities. We verify the performance of the scheme in a multiagent transport simulation and demonstrate that service levels outperform those of the existing rebalancing schemes for identical fleet sizes.
Image Representation of a City and Its Taxi Fleet for End-To-End Learning of Rebalancing Policies
Gächter, Joel; Zanardi, Alessandro; Ruch, Claudio; et al. (2021)
2021 IEEE International Conference on Robotics and Automation (ICRA)
In recent years, mobility on demand has experienced a major revival due to various ride-hailing companies entering the market. Competing in this field requires an efficient operation. Therefore, the applied policy, which cares for vehicle-to-customer assignment and vehicle repositioning, has to achieve good customer service and minimize cost while trying to keep the impact on the environment as low as possible. A promising approach is to coordinate the control of the entire fleet, which is foreseen to become even easier with the possibility of autonomous vehicles in mind. Anticipating future demand requires a good understanding of the spatiotemporal distributions of request origins and destinations, and the resulting imbalance between vehicle demand and availability. This results from a multitude of topological, demographic, and social effects, which are almost impossible to sufficiently capture in a handcrafted model of reasonable complexity. This can be circumvented by leveraging machine learning approaches. In this paper, an image-like representation of the city and its fleet's state is introduced. It is comprehensive and intuitive to use as input to convolutional neural networks, which in the past have already been proven to capture spatial relationships very well. This allows operating on realistic, full-sized traffic networks without greatly increasing the number of parameters the neural network has to learn and, hence, keeps the training effort low. Additionally, this state is combined with a similarly constructed repositioning action, reflecting a 2D distribution of a well-performing operational policy. This approach allows replacement of complex, handcrafted mathematical models by a single, compact, auto-encoder-like neural network.
Simulation-based assessment of parking constraints for automated mobility on demand
Ruch, Claudio; Ehrler, Roman; Hörl, Sebastian; et al. (2021)
Vehicles
In a coordinated mobility-on-demand system, a fleet of vehicles is controlled by a central unit and serves transportation requests in an on-demand fashion. An emerging field of research aims at finding the best way to operate these systems given certain targets, e.g., customer service level or the minimization of fleet distance. In this work, we introduce a new element of fleet operation: the assignment of idle vehicles to a limited set of parking spots. We present two different parking operating policies governing this process and then evaluate them individually and together on different parking space distributions. We show that even for a highly restricted number of available parking spaces, the system can perform quite well, even though the total fleet distance is increased by 20% and waiting time by 10%. With only one parking space available per vehicle, the waiting times can be reduced by 30% with 20% increase in total fleet distance. Our findings suggest that increasing the parking capacity beyond one parking space per vehicle does not bring additional benefits. Finally, we also highlight possible directions for future research such as to find the best distribution of parking spaces for a given mobility-on-demand system and city.
The Value of Coordination in One-Way Mobility-on-Demand Systems
Ruch, Claudio; Richards, Spencer; Frazzoli, Emilio (2020)
IEEE Transactions on Network Science and Engineering
In a one-way mobility-on-demand system or distributed transportation system, customer requests for rides are served by a fleet of agents, e.g., taxis or even autonomous vehicles. We present a simplified three-node network model of such a transportation system in an urban agglomeration. The agents in this model play a non-cooperative game as each one tries to maximize their individual expected profit. We compute Nash equilibria in this game for different customer load cases, specifically the light- and heavy-load cases, and compare the social cost of a system with selfish agents to that of a system with coordinated agents. In particular, we establish a lower bound for the price of anarchy as a function of the system parameters, including taxi fares. We investigate the required mechanism design in the form of the fare ratio for a downtown core node and a city outskirts node that minimizes the social cost caused by selfish agents. Furthermore, we show that this optimal fare ratio is required to bring the social cost for the selfish agents as close as possible to that of the coordinated fleet. The chosen level of abstraction for the network with only three nodes is not intended to accomplish completeness; rather, it provides elementary insights into why mobility-on-demand systems with selfish agents in many cities operate at a sub-optimal level of performance. This paper motivates the investigation of the value of coordination in more complex systems, as well as the study and implementation of coordinated one-way mobility-on-demand transportation systems.
The AI Driving Olympics at NeurIPS 2018
Zilly, Julian; Tani, Jacopo; Considine, Breandan; et al. (2020)
The Springer Series on Challenges in Machine Learning ~ The NeurIPS '18 Competition. From Machine Learning to Intelligent Conversations
Despite recent breakthroughs, the ability of deep learning and reinforcement learning to outperform traditional approaches to control physically embodied robotic agents remains largely unproven. To help bridge this gap, we present the “AI Driving Olympics” (AI-DO), a competition with the objective of evaluating the state of the art in machine learning and artificial intelligence for mobile robotics. Based on the simple and well-specified autonomous driving and navigation environment called “Duckietown,” the AI-DO includes a series of tasks of increasing complexity—from simple lane-following to fleet management. For each task, we provide tools for competitors to use in the form of simulators, logs, code templates, baseline implementations and low-cost access to robotic hardware. We evaluate submissions in simulation online, on standardized hardware environments, and finally at the competition event. The first AI-DO, AI-DO 1, occurred at the Neural Information Processing Systems (NeurIPS) conference in December 2018. In this paper we will describe the AI-DO 1 including the motivation and design objections, the challenges, the provided infrastructure, an overview of the approaches of the top submissions, and a frank assessment of what worked well as well as what needs improvement. The results of AI-DO 1 highlight the need for better benchmarks, which are lacking in robotics, as well as improved mechanisms to bridge the gap between simulation and reality. © Springer Nature Switzerland AG 2020.
Quantifying the Efficiency of Ride Sharing
Ruch, Claudio; Lu, Chengqi; Sieber, Lukas; et al. (2021)
IEEE Transactions on Intelligent Transportation Systems
In unit-capacity mobility-on-demand systems, the vehicles transport only one travel party at a time, whereas in ride-sharing mobility-on-demand systems, a vehicle may transport different travel parties at the same time, e.g., if paths are partially overlapping. One potential benefit of ride sharing is increased system efficiency. However, it is not clear what the trade-offs are between the efficiency gains and the reduction in quality of service. To quantify those trade-offs, an open-source simulation environment is introduced, which is capable of evaluating a large class of operational policies for ride-sharing mobility-on-demand systems. The impact of ride sharing on efficiency and service level is assessed for several benchmark operational policies from the literature and for different transportation scenarios: first a dense urban scenario, then a line-shaped, rural one. Based on the results of these case studies, we find that the efficiency gains in ride sharing are relatively small and potentially hard to justify against quality of service concerns such as reduced convenience, loss of privacy, and higher total travel and drive times. Furthermore, in the assessed scenarios, the relatively low occupancy of the vehicles suggests that smaller vehicles with 4-6 seats, able to handle occasional ride sharing, may be preferable to larger and more expensive vehicles such as minibuses. © 2020 IEEE.

Publications 1 - 10 of 14

Claudio Ruch

Last Name

First Name

ORCID

Organisational unit

Filters

Search Results