Safe Planning in Dynamic Environments Using Conformal Prediction
نویسندگان
چکیده
We propose a framework for planning in unknown dynamic environments with probabilistic safety guarantees using conformal prediction. Particularly, we design model predictive controller (MPC) that uses i) trajectory predictions of the environment, and ii) prediction regions quantifying uncertainty predictions. To obtain regions, use prediction, statistical tool quantification, requires availability offline data – reasonable assumption many applications such as autonomous driving. The are valid, i.e., they hold user-defined probability, so MPC is provably safe. illustrate results self-driving car simulator CARLA at pedestrian-filled intersection. strength our approach compatibility state art predictors, e.g., RNNs LSTMs, while making no assumptions on underlying trajectory-generating distribution. best knowledge, these first provide valid setting.
منابع مشابه
Path Planning in Dynamic Environments
The motion planning problem for mobile robots is typically formulated as follows: given a robot and a description of an environment, plan a path of the robot between two specified locations, which is collision-free and satisfies certain optimization criteria. Traditionally there are two approaches to the problem: Off-line planning, which assumes perfectly known and stable environment, and on-li...
متن کاملAbstract Planning in Dynamic Environments
Solving problems in dynamic and heterogeneous environments where information sources change its format representation and the data stored along the time is a very complex problem. In previous work we have presented a system called MAPWeb (MultiAgent Planning in the Web) that tries to solve those problems by integrating artificial intelligence planning techniques within the MultiAgent framework....
متن کاملSafe and Efficient Navigation in Dynamic Environments
For mobile robots to become ubiquitous, they need to be able to navigate in dynamic environments in a safe and efficient way. This is a challenging problem due to the added time dimension in the search space and the subtle interactions between dynamic agents that are extremely difficult to model. In this thesis, we will address both of these challenges. The challenge of dimensionality is addres...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: IEEE robotics and automation letters
سال: 2023
ISSN: ['2377-3766']
DOI: https://doi.org/10.1109/lra.2023.3292071