Project description
An automatic driving software suite for driverless cars
Self-driving cars need to make decisions based on their mission, rules-of-the-road and information about surrounding objects. This requires an autonomous driving software suite. The tech SME Embotech AG has designed such a suite for defining a constantly updated dynamic driving trajectory. The Embotech Motion Planner (EMP) uses a deterministic approach and physics-based algorithms so that the autonomous vehicle can safely perform complex manoeuvres, as well as high-speed driving, in real time. The aim of the EU-funded EMP project is to ultimately push the transformation from conventional driving to autonomous driving. This will reduce the number of traffic accidents, since the majority are attributed to driver errors.
Objective
More than 90% of road accidents are attributed to driver errors. Thus, there is an urgent need to transform conventional driving towards autonomous driving and make car traffic safer. One of the major challenges in the autonomous car industry, besides sensing, localization and mapping is safe planning of the vehicle motion. Motion planning algorithms are responsible for defining a path to follow, avoiding obstacles and generating the best trajectory that ensures safety, comfort and efficiency. This is a hot topic for the development community, which is still awaiting reliable and robust solutions to solve complex situations, without the need to create pre-programmed solutions for each scenario. To this aim, embotech, introduces the EMP (Embotech Motion Planner), a cutting-edge motion planning software for autonomous vehicles. Using a deterministic approach and physics-based models, the proposed solution generates high-dimensional motion trajectories to assist various self-driving tasks such as valet parking, highway driving and urban driving. The EMP enables fully flexible motion planning, capable to safely perform complex manoeuvres in real time with standard automotive ECUs. Characterized by seamless integration, EMP is one core solution easily adaptable to any new car model. Embotech is a spin-off from the Automatic Control Laboratory from ETH Zurich; our team includes experts in embedded optimizations technologies. We are part of several automotive industry-led incubation programs benefitting from expert coaching and mentoring. Our plan is to reach the market through direct sales to OEMs and Tier 1 suppliers. With a volume-based royalty business model, we expect to reach a SME Instrument Phase 2 project ROI of 8.68 by 2025.
Fields of science
- engineering and technologymechanical engineeringvehicle engineeringautomotive engineeringautonomous vehicles
- natural sciencescomputer and information sciencessoftware
- engineering and technologyelectrical engineering, electronic engineering, information engineeringelectronic engineeringsensors
- social sciencessocial geographytransport
- engineering and technologyenvironmental engineeringenergy and fuels
Programme(s)
Funding Scheme
SME-1 - SME instrument phase 1Coordinator
8005 Zurich
Switzerland
The organization defined itself as SME (small and medium-sized enterprise) at the time the Grant Agreement was signed.