Nikos Aréchiga

• PhD
• Researcher at Toyota Motor North America, InfoTech Labs

Widespread adoption of autonomous cars will require greater confidence in their safety than is currently possible. Certified control is a new safety architecture whose goal is two-fold: to achieve a very high level of safety, and to provide a framework for justifiable confidence in that safety. The key idea is a runtime monitor that acts, along wit...

This paper presents a technique, named stlcg, to compute the quantitative semantics of Signal Temporal Logic (STL) formulas using computation graphs. stlcg provides a platform which enables the incorporation of logical specifications into robotics problems that benefit from gradient-based solutions. Specifically, STL is a powerful and expressive fo...

This paper presents a technique, named STLCG, to compute the quantitative semantics of Signal Temporal Logic (STL) formulas using computation graphs. STLCG provides a platform which enables the incorporation of logical specifications into robotics problems that benefit from gradient-based solutions. Specifically, STL is a powerful and expressive fo...

Real-world large-scale datasets are heteroskedastic and imbalanced -- labels have varying levels of uncertainty and label distributions are long-tailed. Heteroskedasticity and imbalance challenge deep learning algorithms due to the difficulty of distinguishing among mislabeled, ambiguous, and rare examples. Addressing heteroskedasticity and imbalan...

We consider the problem of using reinforcement learning to train adversarial agents for automatic testing and falsification of cyberphysical systems, such as autonomous vehicles, robots, and airplanes. In order to produce useful agents, however, it is useful to be able to control the degree of adversariality by specifying rules that an agent must f...

We describe the concept of logical scaffolds, which can be used to improve the quality of software that relies on AI components. We explain how some of the existing ideas on runtime monitors for perception systems can be seen as a specific instance of logical scaffolds. Furthermore, we describe how logical scaffolds may be useful for improving AI p...

We formulate numerically-robust inductive proof rules for unbounded stability and safety properties of continuous dynamical systems. These induction rules robustify standard notions of Lyapunov functions and barrier certificates so that they can tolerate small numerical errors. In this way, numerically-driven decision procedures can establish a sou...

Deep learning algorithms can fare poorly when the training dataset suffers from heavy class-imbalance but the testing criterion requires good generalization on less frequent classes. We design two novel methods to improve performance in such scenarios. First, we propose a theoretically-principled label-distribution-aware margin (LDAM) loss motivate...

Reliability and safety are important properties in the development of complex cyber-physical systems such as autonomous vehicles. Achieving a reliable autonomous vehicle is a challenging problem, as the unpredictability of the envi- ronment demands a reliable design methodology. Additionally, current testing procedures for ADAS features on vehicles...

We propose a new abstraction refinement procedure based on machine learning to improve the performance of nonlinear constraint solving algorithms on large-scale problems. The proposed approach decomposes the original set of constraints into smaller subsets, and uses learning algorithms to propose sequences of abstractions that take the form of conj...

Some industrial systems are difficult to formally verify due to their large scale. In particular, the widespread use of lookup tables in embedded systems across diverse industries, such as aeronautics and automotive systems, create a critical obstacle to the scalability of formal verification. This paper presents Osiris, a tool that automatically c...

Real-world applications often include large, empirically defined discrete-valued functions. When proving properties about these applications, the proof naturally breaks into one case per entry in the first function reached, and again into one case per entry in the next function, and continues splitting. This splitting yields a combinatorial explosi...

This work addresses the problem of scalable constraint solving. Our technique combines traditional constraint-solving approaches with machine learning techniques to propose abstractions that simplify the problem. First, we use a collection of heuristics to learn sets of constraints that may be well abstracted as a single, simpler constraint. Next,...

Today's automotive industry is making a bold move to equip vehicles with intelligent driver assistance features. A modern automobile is now equipped with a powerful computing platform to run multiple machine learning algorithms for environment perception (e.g., pedestrian detection) and motion control (e.g., vehicle stabilization). These machine le...

Complex systems, such as modern advanced driver assistance systems (ADAS), consist of many interacting components. The number of options promises considerable flexibility for configuring systems with many cost-performance-value tradeoffs; however the potential unique configurations are exponentially many prohibiting a build-test-fix approach. Inste...

This paper is a tutorial on how to model hybrid systems as hybrid programs in differential dynamic logic and how to prove complex properties about these complex hybrid systems in KeYmaera, an automatic and interactive formal verification tool for hybrid systems. Hybrid systems can model highly nontrivial controllers of physical plants, whose behavi...

The use of deductive techniques, such as theorem provers, has several advantages in safety verification of hybrid systems. There is often a gap, however, between the type of assistance that a theorem prover requires to make progress on a proof task and the assistance that a system designer is able to provide. To address this deficiency we present a...

The use of deductive techniques, such as theorem provers, has several advantages in safety verification of hybrid sys- tems; however, state-of-the-art theorem provers require ex- tensive manual intervention. Furthermore, there is often a gap between the type of assistance that a theorem prover requires to make progress on a proof task and the assis...

This paper concerns the use of formal methods to design controllers for dynamic systems such that the closed-loop system satisfies given safety specifications. The usual approach to using formal methods for control applications is to verify safety for an abstraction of the closed-loop system using a candidate controller. We propose an alternative a...

Lyapunov functions are used to prove stability and to obtain performance bounds on system behaviors for nonlinear and hybrid dynamical systems, but discovering Lyapunov functions is a difficult task in general. We present a technique for discovering Lyapunov functions and barrier certificates for nonlinear and hybrid dynamical systems using a searc...

This paper is a tutorial on how to model and prove com-plex properties of complex hybrid systems in KeYmaera, an automatic and interactive formal verification tool for hybrid systems. Hybrid sys-tems can model highly nontrivial controllers of physical plants, whose behaviors are often safety critical. Formal methods can help design sys-tems that wo...

This paper presents a new approach for leveraging the power of theorem provers for formal verification to provide sufficient conditions that can be checked on embedded control designs. Theorem provers are often most efficient when using generic models that abstract away many of the controller details, but with these abstract models very general con...

This paper describes the architecture and implementation of a distributed autonomous gardening system with applications in urban/indoor precision agriculture. The garden is a mesh network of robots and plants. The gardening robots are mobile manipulators with an eye-in-hand camera. They are capable of locating plants in the garden, watering them, a...

This paper describes the architecture and implementation of a distributed autonomous gardening system. The garden is a mesh network of robots and plants. The gardening robots are mobile manipulators with an eye-in-hand camera. They are capable of locating plants in the garden, watering them, and locating and grasping fruit. The plants are potted ch...