Need Exists for Standard ‘Verification Methodology’ in Self-Driving Cars, Says SA

A “common verification methodology” is needed in autonomous-driving development, blogged Strategy Analytics Monday. Current technologies are “barely” meeting the challenge, it said. “Technical model methods provide an effective paradigm for validation since they can provide a mathematical model as a guarantee of correctness,” it said. But that “could be totally different when one is trying to verify a whole autonomous driving platform including sensor fusion, data fusion, sensor calibration and autonomous driving scenario simulation verification,” it said. The announcement from Israeli startup Foretellix opening its “measurable scenario description language” verification protocol to the autonomous-driving “ecosystem” and promoting it as a global standard was a positive step, said SA. The initiatives will help top-tier automakers “use a common, human readable, high level language to simplify the capture, reuse and sharing of scenarios,” it said. It also will enable them to “easily specify any mix of scenarios and operating conditions to identify previously unknown hazardous edge cases,” it said.